Community Identity and Archaeology: Dynamic Communities at Aphrodisias and Beycesultan

Citation preview

Page ix → Page x → Page 1 → CHAPTER ONE

Introduction Why Communities? The “community” is a concept central to human society yet frustratingly difficult to define. It can mean the inhabitants of a village or a neighborhood or the members of a profession, a religion, or a political lobby group. Ethnicity, sexuality, and sporting allegiance are among the many different factors that can be at its root; and the neighborhood watch, the school fair, and the Internet forum are among the numerous ways it can be manifested. Communities, it seems, are everywhere and can potentially relate to almost anything. The term community is also one of the most common buzzwords of the late twentieth and early twenty-first centuries. Like all the best buzzwords, it has come to have multiple meanings, each laden with its own sense of public sentiment and morality. Now, at the end of the first decade of the twenty-first century, community has come to stand for two opposite ideas simultaneously: social responsibility and civic engagement, on the one hand, and potentially subversive ethnic and religious minority groups, on the other. In either of these forms, however, the popularity of the term belies what seems to be a widespread anxiety over modernity and social fragmentation (Delanty 2003, 1). The contemporary anxiety surrounding ideas of the community has manifested itself in many different forms. In 1990, a group of social activists in the United States launched a movement called “Communitarianism,” critiquing what they saw as the individualistic tendencies of contemporary society and explicitly calling for a revival of public morality and social responsibility (Etzioni 1993). The Communitarians, however, are only the tip of a much larger iceberg of social concern. Over the course of the 1990s, a series of surveys showed that people in America and Britain were becoming increasingly anxious about what they perceived to be the “breakdown of the community,”Page 2 → as characterized by social dysfunction and the decay of civic life (Putnam 2000, 25). This popular perception has made the idea of the community a source of valuable political capital. Now, in the early twenty-first century, the term has become laden with distinctly moral overtones and is often connected with “democracy” in contemporary political rhetoric (Williamson, Imbroscio, and Alperowitz 2002, ix-xiv). In Britain, for example, an official governmental post in the cabinet was created specifically to engage with communities in 2005, a move closely followed by the institution of an independent government department for communities and local government in 2006. Over the last two decades, the community has become an emotive symbol for the wellfunctioning society, a panacea for all our social ills. Perhaps paradoxically, the word community has also developed another, slightly less comfortable meaning in the context of post-9/11 national politics. It is now widely used to refer to ethnic or religious minorities within the nation-state. We often hear about the “Muslim community” in Britain, for example, or the “Armenian community” in the United States. In this context, the term implies an uneasy distinction of a subgroup from the main citizen body, hinting at an unspoken agenda or a set of interests that might differ from and potentially threaten those of the central state (Mai Sims 2008; Parekh 2000). These communities are often most in evidence as political forces, lobbying or agitating for a minority cause against the mainstream. This use of the word community as a proxy for the term minority group is a timely reminder that there are always two sides to community cohesion. At its very root, the idea of community is about not just a sense of solidarity but also a distrust of difference. The last two decades, therefore, have seen the concept of the community becoming increasingly politicized in popular discourse (Amit and Rapport 2003, 13ff.; Hoggett 1997). At the same time, the community has also become a common subject within academia. As might be expected, political theorists, sociologists, and anthropologists have all engaged closely with the term in recent years, drawing from and commenting on the popular and political discourse. Social psychologists, for example, have established a quantitative measure called the Sense of Community Index, or SCI, and explored the role of social interaction in stimulating this community feeling (Chavis et al. 1986). Similarly, architects and urban designers have sought ways of using the built

environment to enable and encourage community interaction (Katz 1994; Kingsley, McNeely, and Gibson 1997). At the other end of the scale, anthropological studies have examined the many different social rationales lying behind community identity,Page 3 → from the shared professional and socioeconomic interests of ex-mining communities in the north of England (Dawson 2002) to the communities forged by contestation and conflict in urban America (Abu-Lughod 1994). In many different forms and in many different disciplines, the last two decades have seen an explosion of research aimed at shedding light on the intangible sense of “us-ness” shared by social groups. The result of this is that community identity is currently a particularly vibrant area of academic research. Work is focused not just on identifying either the existence or absence of community identity but also on the way that community identity can be formed and re-formed over time; on how it can become more or less salient depending on the specific social context; and on the way that it can be variously rationalized and articulated through words, images, objects, and actions. The term community has also made a recent comeback within the historiographical disciplines. In archaeology in particular, communities are often talked about as social actors in prehistory and protohistory. It is in communities that hierarchies and social differentiation are first constructed, it is communities that trade with each other and engage in cultural interaction, and it is communities that mediate the experience of state emergence. Despite the increasing usage of the term in archaeology, there has not yet been a coherent attempt to define or theorize the community in an archaeologically specific way. Efforts were made to establish an archaeology of communities1 in the early years of the twenty-first century, most notably in the 2000 volume edited by Marcello Canuto and Jason Yaeger. However, these early insights have not yet been fully embraced, nor have they received enough attention in mainstream archaeology. The term therefore continues to be used fairly broadly in archaeological literature, but without a coherent theorization to support it. A consistent archaeological understanding of the community is still yet to be developed, and a comprehensive archaeology of communities is still yet to be established. This book hopes to go some way toward both of these goals. It considers the familiar concept of the community in a new light, before going on to formulate a coherent approach for an archaeology of communities for the first time. It does so through a combination of theoretical discussion and practical exploration using case studies. The second chapter of this book summarizes the ways in which social theory has conceptualized the community, from the first emergence of the term in the late nineteenth century until the present day. In particular, it highlights how the community is now understood as a form of social identity, actively constructed rather than naturally Page 4 →emerging. The third chapter reflects on how archaeologists have approached communities from both theoretical and practical perspectives over the last century, discussing especially the recent attempts to initiate an archaeology of communities and the various critiques of this movement. It is argued that integrating insights from social theory can answer these critiques and that a new archaeological approach to communities can now be developed by explicitly considering “community identity” rather than the community per se. This approach depends on a close and careful definition of what community identity means, as well as a detailed exploration of how community identity is created, maintained, and manifested. The fourth chapter of the book suggests such a definition and attempts this exploration. In its fifth chapter, the book goes on to propose a method for the archaeological investigation of community identity, focusing on how material culture can be used to construct group identities and on the traces it might leave in the archaeological record. In the second half of this book, the proposed method is tested and illustrated through an in-depth treatment of two sites in western Anatolia in the Late Bronze and Iron Ages (hereafter LBA and IA). The sites of Aphrodisias and Beycesultan developed along different but interrelated trajectories over this time, each showing flexible attitudes to community solidarity in a rapidly changing historical context. These case studies both illustrate and also test the theory and method developed in the first part of this book. They also illustrate the benefits of adopting such an approach. Western Anatolia is an ill-understood region, and the LBA and IA represent an ill-understood period of time. Approaching sites here through the lens of community identity will be revealed as a new way to uncover the complexity of the social landscape, a complexity usually lost in conventional narratives. The two halves of the book are therefore designed to be complementary—theory and practice, ideas and application. As is commonly the case in archaeology, neither one makes complete sense without the other, and both are needed in order to develop a mature archaeology of communities.

This book does not offer to its readers a ready-made and complete archaeology of communities. Rather, it aims to open up a wider debate and discussion on the subject. Building on the work of others, the book represents a first attempt to develop a coherent and comprehensive approach to community identity in archaeology. The approach offered here is neither foolproof nor perfect—instead, it is the first formulation of a body of theory that, I hope, will be developed much further in years to come. First, however, it is necessary to consider the range and the scope of the issue at hand,Page 5 → the frame of reference and the context of the debate. The following section provides a brief introduction, discussing how communities have been invented, imagined, and interpreted in LBAIA western Anatolia to date and the implications of this for our historical understanding of the region.

Illustrating the Problem: The Archaeological Invention of Communities in Western Anatolia For decades, archaeology has taken the idea of the community largely for granted. Little explicit theorization has been afforded to the subject, yet communities appear frequently in archaeological writing—both Old World and New World, both empirical and social. Few places is this more in evidence than in western Anatolia. In this region, archaeologists have frequently assumed the existence of communities, invested these invented communities with social meaning, and built complex historical arguments based on their interpretation. This process, while common in archaeological practice more generally, is particularly problematic in western Anatolia. Inventing communities has been crucial in the development of the current historical understanding of the region, an understanding that should perhaps be called into question. The term community is often used unreflectively in archaeological writing. Most commonly, it is assumed that communities are the building blocks of the archaeological landscape. Communities tend to be talked about as if they were social actors—as if the communities were unitary entities in themselves that interacted with each other, developed new social structures, and created different cultural forms. At this scale, internal divisions within the community are ignored, and the actions of individuals or minority groups are not thought relevant. The community is assumed to be a coherent entity—it is treated as an unproblematic and natural unit. In addition to this assumption of the community as a natural unit, communities are often equated with sites. It is frequently assumed that the archaeological site, imagined as a nucleated settlement, is equivalent to the physical remains of this natural social unit. Groups of people living together in close residential proximity, it is supposed, can be considered as an explicitly social collective, acting as a single unified social actor. The common archaeological use of the community is therefore somewhat flawed (see chapter 3 for a fuller discussion of archaeological uses of the term community). From sites, archaeologists commonly infer unitary communities;Page 6 → and from these communities, we tend to build complex historical and social narratives. In archaeology, therefore, communities are rarely explored as concepts in their own right—rather, they make recurring cameo appearances as natural social units or the unproblematic social equivalent of the site. Western Anatolia during the LBA and IA provides a vivid example of this trend. Information about this period in western Anatolia is patchy, with a disrupted archaeological record and very little historical documentation. The region lies at the interface of what are widely considered to be two distinct political and cultural spheres: the central Anatolian plateau to the east and the Aegean to the west (fig. 1). Throughout their long history, the people of western Anatolia have never been a full part of either sphere but have instead maintained strong connections with both. In antiquity, the region is usually characterized as being a border zone—an intermediate area caught between the East and the West, the Aegean and the Anatolian plateau, Europe and the Orient, the classical world and the Near East (seechapter 6 for a fuller discussion of this). This characterization of western Anatolia rests, however, partly on the archaeological invention of communities in the region. In general, western Anatolia has been the object of scholarly interest primarily as a place “in-between,” an area in which to study the interaction of competing and overlapping spheres of influence. Such studies are usually effected by taking communities/sites as a convenient unit of analysis and assessing the extent of Eastern or Western influence on each one. These communities/sites are then assigned places along a cultural spectrum—ranging from the wholly Aegean, through the Aegean-influenced and the plateau-influenced, to the wholly plateau-focused (e.g., Briant 1996, 682-83; Bryce 2006, 77-86; C. Gates 1995; Lloyd 1956, 6ff.; Ma 1999,

43-101; Marchese 1986, 24; Mcqueen 1996, 39; Mee 1998; Neimeier 1999; Seeher 2005). The shifting allegiances of individual communities within this border zone, either eastward or westward, are therefore seen as the means of analyzing border-zone dynamics and of shedding light on the wider interaction between the Aegean and central Anatolia. In LBA-IA western Anatolia, therefore, it has been frequently assumed that the archaeological site is directly equivalent to a socially meaningful community and that this community has a coherent and unitary nature. By making such assumptions, it has become possible to use these communities as passive indicators of external influence and to map the overlap of the central Anatolian and Aegean spheres. The existence of communities in western Anatolia has therefore been widely assumed, and this archaeological invention of communities has contributed to the common characterization of the region as in-between. Page 7 → Page 8 → It is, of course, extremely important to consider the traces of cultural interaction at different sites within LBA-IA western Anatolia. However, the assumption that these sites necessarily constitute communities ignores individual and group agency within these settlements. Community cohesion is seen as automatic, rather than a specific type of social strategy. Individual choice and the actions of minority groups are excluded from analysis, as is the potential for material culture to be selectively adopted and adapted. Instead, the communities/sites are cast as passive recipients of external cultural influences, the extent of which can be measured and cited as evidence for “in-betweenness.” The peripheral nature of western Anatolia is both a starting assumption and a natural conclusion of this process. The invention of communities is therefore both a result of and a contributing factor to the current characterization of western Anatolia in archaeology as a border zone. Assuming the straightforward existence and unity of communities permits us only a narrow understanding of western Anatolia as a region. A broader and more nuanced view can be sought by reassessing the archaeological approach to communities and by investigating the complex and changing dynamics of community identity. While the social dynamics within a settlement may indeed have encouraged a conscious sense of community identity at particular times and in particular historical circumstances, there may equally have been times when the concept of the community was not so important. In addition, individuals and groups resident in the settlement would have struggled to shape this community identity (where it was extant), in different and sometimes contradictory ways. If these points are taken into account, it becomes evident that the material from LBA-IA western Anatolia therefore holds a far greater potential than simply to plot the extent of Eastern and Western cultural influence on a border zone. It can shed light on a much more complex set of social and geographic dynamics, where cultural traits and social practices were actively used by individuals, groups, and communities in response to a rapidly changing social and historical context. Adopting an approach focused on community identity will help to tease out some of the strands of this complexity, offering new perspectives on old questions. First, however, it is necessary to revisit the concept of the community itself to establish what we mean by a community, to define the significance of community identity, and to understand the interweaving links between geographic locality, community identity, and material culture.

Page 9 → CHAPTER TWO

Theorizing the Community Nineteenth-Century Certainties: The Natural Unit The concept of the community is one with a long and rich history. Ideas about communities have been formed and re-formed over the decades, with definitions constantly changing (Bruhn 2005, 29ff.; A. P. Cohen 1985). However, it is possible to trace the first emergence of community as a critical term in social theory back to the late nineteenth century and the work of the German theorist Ferdinand Tönnies. Tönnies is credited with pioneering the academic study of the community, focusing his arguments on human geography and social organization (Chorney 1990). His work is still widely read and relevant today, as it speaks strongly to contemporary concerns over the loss of traditional forms of life and the onward march of modernity. His classic work, Gemeinschaft und Gesellschaft (Community and Civil Society; Tönnies 1887), is still very influential despite being published over 120 years ago. In this book, Tönnies compares close-knit rural communities (Gemeinschaft) with modern urban society (Gesellschaft), arguing that these different forms of human spatial organization also resulted in radically different forms of human psychology and mentality. Tönnies's vision of rural “community” and urban “society” as diametrically opposed forms of social being was heavily value-laden. He explicitly argues that community, or Gemeinschaft, is the preferable form of the two and that it occurred naturally throughout human history. In contrast to this, he stresses the negative effects and mass depersonalization of urban modernity, asserting that society, or Gesellschaft, is a relatively new and inherently unnatural phenomenon. Throughout the book, therefore, Gemeinschaft and Gesellschaft are presented as opposites: unity as opposed to isolation, collective action as opposed to selfish individualism, organic harmony as opposed to mechanical structure,Page 10 → nature as opposed to reason. The two forms of social organization are also seen as historically situated—Gesellschaft was a consequence of recent modernity and, as Tönnies hopefully suggested later in his life, was a phenomenon that might still be turned back. Tönnies's arguments should be seen in the wider context of their time. Germany, like much of Europe, was undergoing rapid industrialization, with urban centers growing at dizzying speeds and technology advancing at a pace beyond all expectation. Anxiety over weakening social bonds and traditional structures was widespread, and Tönnies's work on communities is only one, albeit very direct, manifestation of this concern (Danahay 1993, 20ff.). Tönnies's conception of the community as a naturally occurring, small-scale, close-knit, and rural form of social organization was very influential, both on a theoretical level and for its moral implications. His younger contemporaries built on these foundations, integrating into their work both Tönnies's sense of the inherent morality of the community and his conception of it as being the “natural” form of human social organization. Within a few years of Tönnies's seminal work, the French and German sociologists Émile Durkheim and Georg Simmel were to borrow from it in their own discussions of modernity and community. In Durkheim's theory of organic and mechanical solidarity (1893), the community was a mechanism of traditional social control through solidarity and cohesion, and this cohesion was necessarily weakened in an industrial urban setting. The complex division of labor in a modern urban economy, he argued, inevitably resulted in a relaxation of social bonds and therefore also in the erosion of solidarity and community. Unlike Tönnies, Durkheim did not see this development in a wholly negative light. Instead, he saw the unity of traditional “mechanical” solidarity as potentially restrictive and suggested that its demise might allow for greater human creativity. In contrast, Simmel, who largely agreed with Durkheim in his analysis of modern urban living, came to the opposite conclusion about its moral and social implications (1903). Simmel worried that the weakening of community bonds in cities would eventually lead to a complete breakdown of society, having an irreversible negative effect on the human psyche in the process. Despite their different judgments on the decline of the community, both Simmel and Durkheim followed Tönnies in thinking that such a decline was happening. Both also followed Tönnies in their use of the term community to evoke an almost wistful nostalgia for a form of small-

scale, rural social grouping governed by intimate and familiarPage 11 → social relations. Finally, both followed Tönnies in the view that a sense of community would arise naturally out of a particular set of social relationships and interactions, which were themselves determined by the physical and economic structures around them. The concept of the community as a naturally occurring form of rural social organization was therefore well established by the turn of the twentieth century. Following this, several scholars began to explore the nature of such communities, pioneering the idea of community ethnographies in their fieldwork. Notable among these is the American Charles Josiah Galpin's study on rural communities in Wisconsin (1915). Galpin set out to establish, in practical sociological terms, a definition for the rural community, laying out a set of characteristics and criteria by which communities could be assessed. Like Tönnies, Simmel, and Durkheim before him, Galpin focused on human relationships as the key determining factor of a community—especially the close and frequent relationships of interaction and exchange that he considered to be fundamentally set into a rural locale. Networks of relationships that were close, regular, and personal were considered to be the essential characteristic of the community. Such relationships, Galpin suggested, were the result of repeated interaction and everyday activity, which was itself the result of the geographical and economic structures of rural Wisconsin life. During the late nineteenth century and at the start of the twentieth, the community therefore was considered to be a naturally occurring form of social organization, which had endured for centuries before its disruption by the Industrial Revolution. It was seen as an almost exclusively rural phenomenon and in direct opposition to modern urban living. It was assumed that a sense of community would naturally emerge in a traditional rural setting with a small-scale economy and limited geographic spread and would require no conscious ascription of identity on the part of community members. Community therefore became largely synonymous with traditional rural society and was used as a kind of easy shorthand for dominant forms of social relationships that might be expected within such a society. Central among these were relationships of intimacy and reciprocity that were thought to be absent in modern city life. Relationships like these—close, personal, and familiar—were thought to arise automatically within a community and also to be largely dependent on community organization for their existence. For thinkers in this period, the community simply was—traditional, rural, and, most important, completely natural. Page 12 →

Mid-Twentieth-Century Structures: The Ecological Unit This definition of community as a naturally occurring form of social organization persisted well into the early and even the middle twentieth century. However, it was given a sharper form and focus in the work of the Chicago school of social geography (Bulmer 1984; Hannerz 1980). Under this school of thought, the physical environment was considered to be a determining factor just as much as in the work of Tönnies and the other early community theorists. Unlike their predecessors, however, the theorists of the Chicago school shifted their focus from the rural to the urban, often carrying out their research in the rapidly growing city of Chicago itself. They argued that while traditional communities may not have the same forms in urban settings as in rural ones, communities nonetheless did also exist in cities, with their own types of structures and dynamics. What followed were the first studies of communities as modern, dynamic, and changeable social forms. A crucial factor influencing this shift of thought was the increasing scale of migration and urbanization, especially in the United States. Also significant was the experience of the First World War, which disrupted traditional social systems across Europe and stimulated a widespread rethink of traditional moralities and lifestyles in both Europe and America. Modernity was no longer viewed with late nineteenth-century suspicion but embraced with early twentieth-century optimism. Members of the Chicago school built on the environmental determinism of earlier thinkers by integrating ideas from ecology and evolutionary biology into their work. Drawing from biological comparisons with organisms, they thought that social forms would also adapt to the environment where they were situated, and school members worked to trace the adaptation of community organization in the urban environment. Each city was viewed as being divided into a mosaic of separate ecological zones, with different groups segregating themselves into ethnic and religious communities both spatially and socially (e.g., Park, Burgess, and MacKenzie 1925; Whyte 1943; Zorbaugh 1929). It was recognized that individuals were often part of several of these urban communities at the

same time—during work, at home, at school, at the local dance hall, and so on. Urban communities were therefore characterized as being more complex than their rural counterparts. First, they were densely micropat terned in spatial terms, sometimes even overlapping with adjacent neighborhoods. Second, their membership was more flexible, with individualsPage 13 → belonging to more than one community at different times in their everyday routines. Third, the social interaction within these communities was not always intimate and direct (primary relationships) but could often be impersonal and indirect (secondary relationships). The combination of these factors, members of the school argued, necessarily led to a greater sense of fragmentation and some weakening of community bonds. However, following Durkheim's perspective, this was seen as having positive consequences, such as increased freedom and creative license (Park, Burgess, and MacKenzie 1925, 40-41), as well as negative ones, such as social disorder, personal isolation, and delinquency (Wirth 1938, 23). Overall, the Chicago school argued that urban communities were complex and qualitatively different from traditional rural communities but were still communities nonetheless. The conception of what qualified as a community had changed. Communities were no longer considered to be traditional or monolithic, exclusively tied to certain types of living arrangements and resulting in certain types of social relationships. However, it was still thought that external observers could identify a community using objective methods, specifically if they could discern a coherent spatial focus and a reasonable level of regular social interaction. In the decades after the pioneering early work of the Chicago school, the practice of “community studies” became increasing popular in the social sciences on both sides of the Atlantic (Bell and Newby 1971, 1974; Frankenburg 2004). Involving the examination of social relationships set within a particular locality, such studies often resulted in richly detailed descriptions of social life in specific areas. The locations of these studies varied—villages, urban neighborhoods, and small towns were all focuses for different community studies in the 1940s through the 1960s. This genre of academic writing spanned sociology, geography, and anthropology and gave rise to many “classic” case studies. These include Williams's study of the English village of Gosforth (1956); Seeley, Sim, and Loose ley's work on the wealthy Crestwood Heights suburb of Toronto (1956); and Young and Wilmott's research on the deprived Bethnal Green area in London (1957). While community studies such as these provided a wealth of detail about their respective subjects, they often employed the term community without reflection. The word community was straightforwardly used to refer to the population living within a particular locale, and it carried almost no connotations at all of identity or conscious association. By the mid-twentieth century, therefore, the community had come to be viewed as an ecological structure—an objective designation for a population of organisms living and interacting within a specific ecological niche. Page 14 → The practice of community studies came under fierce criticism in the late 1960s for precisely this reason—the lack of a critically theorized concept of the community itself. This was strongly condemned by Margaret Stacey in 1969 in her belligerent article “The Myth of Community Studies.” In it, she attacked both the vagueness of her contemporaries when it came to defining the community and also their unquestioning assumption of the community as a natural social structure. She also argued that the idea of the community served no useful purpose in the social sciences, claiming that “in so far as so-called community studies have been based on an uncritical structural-functional approach they are not defensible” (138). Stacey identified two elements of communities that she did not consider to be analytically valid: first, the concept of situating a community in a given geographical area; and second, a sense of community belonging. Neither, she asserted, could be defined with any accuracy, and there was no apparent necessity for the one to be linked with the other. In place of the term community, Stacey suggested that scholars instead referred to “local social systems,” as a more neutral term that avoided emotive connotations. Stacey's critique was to signal the death knell for traditional community studies (Hoggett 1997, 6), consigning it “for some time into an abyss of theoretical sterility” (A. P. Cohen 1985, 38). While some research on local economic systems persisted, it was not until the mid-1980s that there was a radical rethinking of the concept and that the community was reinstated as a mainstream focus for academic study.

Late Twentieth-Century Revisions: The Socially Constructed Unit Anthony Cohen's seminal work of 1985, The Symbolic Construction of Community, revived the academic study of

communities after the stagnation of the 1970s. During the early 1980s, Cohen developed a new theory of the community that radically revised the existing approaches. He argued that a community was not a natural social unit arising automatically from residential proximity and locality but, rather, a dynamic and socially constructed identity group (1982). For Cohen, the community was not so much a structural unit as a locus of identity, a way of articulating a collective sense of “us” in relation to an external “them.” Comprehensively deconstructing earlier approaches to the community, Cohen drew heavily from Barth's ideas on boundaries and identity to develop a theory of community as a social construct (A. P. Cohen 2002, 166). The structures of community organization, he argued, had dominated the academic discourse for too long. Like Stacey before him, Cohen criticized the previous assumption that structures and environmental conditions would necessarily govern meaning and social perception (1985, 70ft). Instead,Page 15 → he argued that the community was essentially a mental construct—an idea in the minds of its members (1985, 97ft). This idea, he suggested, was articulated and symbolized through the active manipulation of cultural forms. Previous work on communities had highlighted the relationships between community members, assuming that different structural conditions would result in different degrees of familiarity between individuals. Cohen, by contrast, was less concerned about the interaction between members and focused instead on the relationship between insiders and outsiders (1982, 293ft). The definition of the community vis-à-vis non-members was the crucial consideration, he argued, and he suggested that culture was the means by which this distinction was made. Because of this, Cohen encouraged a research focus on culture rather than structure, because culture was the way in which people developed an understanding of self as opposed to other, of “us” in relation to “them.” Community, Cohen argued, was constructed symbolically through social and cultural practice and could only be maintained through constant social and cultural negotiation. Cohen's explicit new theorization of the community complemented Benedict Anderson's contemporary idea of the “imagined community” (Anderson 1983). Anderson also viewed the community as a mental construct, and he focused on nationalism in this context, arguing that national identity was a sense of collective identity created through social and cultural practice. Anderson suggested that since a community was essentially an imagined entity, it was not strictly necessary for there to be any direct interaction between its members at all. While early community theorists had defined communities as natural sociogeographic units with close, face-to-face interaction, the Chicago school and the community scholars of the mid-twentieth century had argued that communities were organic sociogeographic units held together by a combination of direct and indirect social interaction. Anderson's suggestion was even more radical: communities as imagined sociogeographic units where no interaction was strictly necessary and where the only important elements were the concept of the community itself and the cultural forms that symbolized it. Under Anderson's formulation, the structural aspects of community became all but irrelevant, and what took its place instead were the cultural mechanisms by which the sense of a collective “us” was produced.Page 16 → The question of how community members related to each other, whether personally or impersonally, whether with familiarity or distance, was a question that had dominated the academic study of communities from Tönnies onward. Now, however, the focus shifted to how community members interacted not with each other but with cultural forms. Cohen and Anderson's work dramatically recast academic understandings of the community. The community was no longer considered as emerging naturally from certain living conditions or structural arrangements. It was now theorized as being an imagined or mental construct, created in social practice through the deployment of cultural forms that stimulated a sense of solidarity between members and articulated a sense of difference from outsiders. Both Cohen and Anderson stressed the importance of culture and social practice, thereby setting the agenda for the academic study of communities until the present day. Anderson and Cohen's books roughly coincided with the popularization of the community in political discourse. As already discussed in chapter 1 of this book, the late 1980s saw increasing interest in the community among political and social commentators. This growing visibility of the community in the public eye, coupled with the new theoretical insights from Cohen and Anderson, resulted in a revitalization of the subject within academia. There was considerable new interest in exploring how a sense of community came into being. As Gupta and Ferguson put it, “Instead of assuming the autonomy of the primeval community, we need to examine how it was formed as a community out of the interconnected space that always already existed” (1992, 8; original emphasis).

Research into diaspora communities and migrants became especially popular, as these could be studied for the ways in which newly formed communities established themselves and created a sense of place and collective belonging in new environments (e.g., Malkki 1992). Explicitly politicized research into the community also appeared, considering the more practical consequences of this idea of the community (see papers in Hoggett 1997). These new approaches to community studies ranged from recommendations concerning public architecture of urban housing estates (Kingsley, McNeely, and Gibson 1997) and psychological measurements for community spirit (Chavis et al. 1986) to minorities articulating a sense of community through literature (Bhabha 1994, 328ft) and multiethnic communities forging their collective identities through confrontation and competition (AbuLughod 1994). Cohen and Anderson's ideas were to lead, however, to some problematicPage 17 → eventual conclusions. In the imagined, socially constructed community, interaction between the community's members was no longer necessary—only interaction with ideas and cultural forms was essential. With the advent of printing, the mass media, and eventually the Internet, it was pointed out that the idea of an imagined community could be communicated remotely and possibly even without any contact or interaction between community members at all. This opened up possibilities for considering communities on a previously unimagined geographical scale—as in the case of Anderson's national communities—or even as divorced from geography completely. Social theory of the later 1990s shows a growing interest in changing perceptions of space, virtual landscapes, deterritorialization, and imaginary geographies, taking its cue from the deplaced experience of contemporary modernity (e.g., Appadurai 1996; Bahrani 1998; Robertson 1995). These wider developments contributed an additional facet to the new theory of community: the community not tied to any real physical location at all. A wide range of different social groups now came to be studied under the banner of communities, including communities in cyberspace, communities of practice, religious communities, sporting communities, the gay community, the scientific community, and even that most disparate community of all, the academic community. The deterritorialization of the community has led to the current theoretical situation as described in chapter 1. Communities, as they are currently understood, can potentially be everywhere and can potentially relate to anything.

Socially Constructed Geographic Communities As has already been seen, the word community can be applied, in current usage, to a seemingly infinite number of social phenomena—any group sharing a common sense of identity and interests can be considered a community, and the term can be used almost interchangeably with the term social group. It seems that the same ideas that initiated an exciting new approach to communities in the 1980s have ultimately led to challenging new problems for community theory in the late 1990s and 2000s. The community has now come to be seen as purely an imagined mental construct, not necessarily involving any social interaction between members at all (Amit and Rapport 2002, 17; Delanty 2003, 3). Recent critiques have bemoaned the fact that this theorization has led to an unhelpful broadening of the term so that it has become almost useless as an analytical category. Given this impossible taxonomic situation, it is unsurprising thatPage 18 → theorists in the early twenty-first century have sought to shift the discussion in one of two directions: either ignoring the issue of definition completely or redefining the community even more closely. Some scholars of community have opted to ignore the extremely broad usage of the term community altogether and now focus instead on how communities are given social meaning. They are researching not how communities are imagined on an intellectual level but how they are constructed in a practical sense through culture and lived experience. Recent work has stressed that while all communities might fundamentally be mental constructs, a community must be realized to have any kind of social impact. After all, as Vered Amit has commented, “if communities must be imagined, then by the same token, what is imagined can only be truly felt and claimed by its potential members if they are able to realize it socially” (2002, 8). The community that is nothing more than an intellectual construct is no longer of central interest, and attention has shifted instead to those communities that have some capacity as a social force. The emphasis is now on the social mechanisms that create a sense of community, rather than on the intellectual constructs that define it. How does a community become realized, and how is it maintained through social practice? Why does a community identity become salient at some points in time and less important again in others? This recent trend marks a dramatic shift in the academic study of the

community and a qualified return to the research questions of the early and middle twentieth century. The end of the twentieth century saw a concentration on the conceptual boundaries drawn between one community and the next—on the construction of the external Other. At the start of the twenty-first century, it was felt that the balance had swung too far in this direction and that more emphasis should also be placed on the sense of community shared between members—on the construction of a sense of “us” or self (B. Cohen 2000; Jenkins 2004, 108). This renewed interest in the relationships between community members harks back to the insights of some of the early community studies. However, these older studies, which tended to assume that community members would automatically have a sense of “us,” concentrated on assessing the quality of direct and indirect interaction. The most recent research, by contrast, does not assume a conscious sense of “us” and instead investigates how this sense can be constructed, negotiated, and contested through social and cultural practices. While one twenty-first century movement has sought to avoid defining the community altogether, a contemporary trend has attempted to find more specific definitions for particular types of communities.Page 19 → Through this closer definition of the different types of communities, it is hoped that comparison of like with like will be possible. In this new scheme, communities in the broad and generic sense are now often referred to as relational communities. A rich literature is available concerning different types of relational communities, exploring communities as diverse as diasporas, virtual communities, and political pressure groups. But perhaps more relevant to the discipline of archaeology in general and this book in particular is the designation of one particular type of community—the geographic community. Geographic communities are rooted in a particular locality, such as a neighborhood or a village, and have spatial correlates as well as the social ones that all relational communities have. It is widely recognized that this particular type of community is qualitatively different from other types of community that are more dispersed, and studies of geographic communities have come to the fore in the last decade (e.g., Gray 2002; Kempney 2002; see papers in Lovell 1998). The geographic community is a subset of the wider category of relational communities. Like other relational communities, it is a mental construct where members of the community feel a sense of cohesion and shared identity based on some perception of commonality. However, unlike other relational communities, this commonality focuses primarily on shared space and location—a village, a neighborhood, or a locality. There may be a number of subsidiary factors that also contribute to the geographical community's identity, such as descent or ethnicity, religious faith, profession, language, and nationality. However, all of these factors are built on the essential foundation of the community's basic sense of commonality, which is shared place and the shared experience of residential proximity. Emplacement and rootedness in space is a central part of the common experience. In other words, members of a geographic community must live together (Bell and Newby 1971, 21-32). It is important to stress that the inhabitants of a village, a settlement, or a neighborhood do not necessarily constitute a geographic community. Indeed, people living within a locality do not necessarily have to think of themselves as a community at all. For the inhabitants of an area to form a community, they must share a social identity as well as a geographic space, and they must consciously see themselves as a social collective. They are not just the unconnected residents of the same place—instead, they must act as a relational community set within a geographic locale. Crucially, they must have a conscious sense of community identity. Practice theory and phenomenology can shed light on how this conscious sense of identity can arise from residential proximity (Bourdieu 1977; Heidegger 1927; Rohrer 2007).Page 20 → While a sense of collective identity does not occur automatically from simple geographic proximity, it can potentially crystallize through the embodied experience of coresidence and shared social practices (A. P. Cohen 1985; Dawson 2002; Peters 1992). This can be illustrated by a host of modern examples demonstrating that the experience of living together or of using shared community spaces and undertaking shared community activities can foster a sense of shared belonging (Karp 1997; Schorr 1997). Residential proximity therefore allows for but does not necessarily entail a high level of regular and personal interaction between members. The phenomenological and social experience of this interaction can form the basis on which a conscious sense of community can be constructed. It is important to note that residential proximity does not explain how and why, in any given example, a conscious sense of community may be constructed on this foundation—it merely creates the environmental conditions where such an identity can

become salient. The answers to the “how” and “why” questions are dependent both on the place and on the changing historical circumstances. Answering them involves examining the construction, contestation, and maintenance of a sense of community. Two criteria therefore emerge for defining a geographical community. First, it must be set within a certain geographic area and have a certain spatial focus. Second, the members of this community must consciously identify themselves with that place and each other, ascribing to a sense of collective identity. A geographic community therefore must be an imagined community, a realized community, and an emplaced community all at once. The latter two of these characteristics make the geographical community an appropriate subject for archaeology. Through material culture, it may be possible to examine the realization of the intangible mental construct of community in social practice. Through a spatial and geographic study, it may also be possible to consider the emplaced nature of the community. Archaeology is therefore a discipline well suited to the study of geographic communities. This is discussed in more detail in chapter 4. Today, our understanding of the geographic community builds on the insights of over a century of theoretical and practical research. It combines the focus of the early theorists on interaction with the more recent recognition of identity as a social construction, blending the old interest in place and location with the new approaches of practice theory and phenomenology. The geographic community is no longer thought to be a natural social structure, emerging automatically from environmental conditions.Page 21 → Nor is it considered to be a purely intellectual concept, defined wholly by individual decisions and choice. Rather, it is now understood as a conscious mental construct, built both on and through social practice and lived experience, which is itself facilitated by residential proximity and regular direct interaction. The intellectual, the environmental, and the social all have their part to play in its construction. It is the last of these three factors—the social—that offers up the most tantalizing research questions for archaeologists. How and why is a sense of community identity constructed? Chapter 4 of this book attempts to address these questions, considering the dynamics of community identity. Chapter 5 then sketches out a tentative methodology for approaching these questions in archaeology using material culture. First, however, it is important to consider how archaeologists have approached the community in their work so far.

Page 22 → CHAPTER THREE

Communities in Archaeology Archaeology has never engaged very closely with the idea of the community. When it comes to the conceptualization of the community, archaeologists have largely followed the broad trends in social theory outlined in the previous chapter. In the early days of the discipline, archaeologists assumed that the community was a naturally occurring social unit, directly equitable with the site or the settlement. Later, in the mid-twentieth century, there was a tendency to consider communities as organizational structures, ranged along a rising scale of human complexity. The idea of the community then fell out of fashion in archaeology, roughly in parallel with the rejection of community studies in sociology, and the use of the term came under damning scrutiny and criticism in both disciplines around this time. Unlike in other social sciences however, archaeologists remained wary of the community until the start of the twenty-first century, when the first moves were made toward establishing a theoretically sensitive archaeology of communities. To date, this has met with limited success, and the emergent archaeology of communities has attracted as much criticism as it has praise. I maintain in this book that the critics' main concerns can be addressed by a closer and more nuanced understanding of the concept of the community itself, the basics of which I have outlined in the previous chapter. The central problem is that despite remaining in step with developments in social theory during the early and middle twentieth century, archaeology has yet to take on the most recent theoretical insights regarding communities.

Archaeological Approaches to the Community In the late nineteenth and early twentieth centuries, the approach of the culture history school dominated mainstream archaeology (Jones 1997, 15ff.; Trigger 1989, 148ft).Page 23 → Its focus was on the identification and interplay between different cultures. These were thought to be internally coherent and biologically selfreproducing population groups, inhabiting a fixed geographical area and recognizable by standard recurring cultural traits. Under the culture history school, archaeologists sought to identify social groups with a territorial extent on the basis of the distribution of material culture styles. This program was most famously laid out by V. Gordon Childe in the preface to his 1929 synthesis of Danubian prehistory. We find certain types of remains—pots, implements, ornaments, burial rites, house forms—constantly recurring together. Such a complex of regularly associated traits we shall term a “cultural group” or “culture.” We assume that such a complex is the material expression of what would to-day be called a “people.” (1-2) Under the culture history school, individual sites became the main unit of analysis. Sites themselves were usually assumed to be nucleated settlements. Archaeologists looked at whole sites to judge whether they could see the recurring cultural assemblages and “regularly associated traits” that were thought to be the hallmarks of a given culture. The site therefore not only became the main unit of analysis but was also considered to be the main political and social unit for the population that lived there. While the word site was therefore used to refer to the material remains and geographic extent of the settlement, the word community came to be used for its human occupants. In the formative days of archaeology, therefore, it was widely assumed that the community was merely the human correlate of the site. In this, archaeologists were following the then-prevailing view of communities as natural social units, automatically arising from close living arrangements and the sharing of space. This approach assumed that there was little or no social differentiation within the site and that the community could be treated as a single, unproblematic actor in the larger and more important drama played out between cultures. Indeed, archaeologists confidently wrote about the community as if it were an independent and coherent social actor. “A village community might be glad enough to obtain luxury articles, such as shell, from strangers by barter,” suggested Childe, “but still shrink from becoming dependent on such traffic for raw materials”; Childe was speaking of evidence for trade at Neolithic village sites in western Europe (1958, 76). The communities discussed

in the archaeological literature of this period are not communities as we would understand them now in the geographical sense: social groups with a conscious sense of solidarity based on, among other things, residential proximity. Neither are they communities as we would Page 24 →understand them now in the relational sense: imagined collectives who perceive a common sense of interest and identity. Rather, community is used as a socially empty term—it refers only to the inhabitants of a given site. The culture history approach was challenged in the mid-twentieth century by a new school of archaeology, which preached a gospel of archaeology as scientific anthropology. The New Archaeology, or processual archaeology, argued vociferously against the concept of archaeological cultures (Binford 1962, 1968). In line with contemporary social theory, the New Archaeology criticized the idea of monolithic cultures with fixed social and geographic boundaries. Instead, it focused on the way social systems are composed and generated, looking inward at the internal dynamics within societies, at individual identities such as gender, status, or rank (Trigger 1989, 386ft). While the culture history school had therefore focused on cultures at a macroscale, seeing communities merely as pixels making up the overall image, the New Archaeology focused on social systems at a microscale, seeing communities in terms of their constituent elements and processes. Systems, structures, and processes were key in this new movement, which encouraged the use of quantitative methods and placed a high value on objectivity, calculation, and statistics. It is therefore unsurprising that the term community took on more of a structuralist meaning during this period, to refer to a certain type of organizational unit. Under the New Archaeology, different types of social organization were seen as ranked along a socioevolutionary scale, ranging from the simplest to the most complex. The community was accorded a specific stage on this scale somewhere between the household and the chiefdom (e.g., Johnson and Earle 1987, 123ft; Trigger 1968). In one influential book, for example, the second chapter deals with “Analysis at the Household Level,” the third with “Analysis at the Community Level,” and the fourth with “The Village and Its Catchment Area” (Flannery 1976). After the “politically egalitarian” societies of deep prehistory, it was assumed that local communities formed the basis for the simplest forms of government and that the existence of either Gemeinschaft or Gesellschaft depended primarily on the administrative system of the settlement (Layton 1972). In archaeology, therefore, the community came to be understood as a natural organizational structure—a similar conception to that of the Chicago school and contemporary social geography. Another similarity between archaeological and Page 25 →sociological approaches to the community at the time was the ecological perspective. It was believed that “the maximum size and stability of a community are quite obviously limited by the environment” and that “ecological factors play an important role in determining the major types of community pattern” (Trigger 1968, 61). For archaeologists in the mid-twentieth century, the community was neither a form of socially constructed identity, as in current thinking, nor a natural social unit, as in the work of the first community theorists. Instead, it was a unit of societal structure. This perspective on the community largely conformed to the contemporary view of the time, when communities were seen as ecological structures that could be quantifiably analyzed and discussed (see chapter 2). The community could be identified as a particular type of component in the system of wider human society. Alongside this structural use of the term community, its traditional use to refer to the human inhabitants of a site also persisted. Nowhere was this more common than in settlement archaeology—an area fast growing in popularity at the time. This agenda is laid out by Chang: “In practice…the settlement—the local context wherein the community is presumed to have resided and gone about its daily business—must substitute for the community” (1968, 3). To date, archaeology has therefore tended to use the term community in one of two equally problematic ways: first, as interchangeable with site and viewing the community as a natural unit; and second, as a structural or evolutionary term, viewing the community as an organizational unit. Both of these usages drew from contemporary theorizations of the community and are very much of their time. However, both usages also continue in archaeological literature today. The structuralist use of the term appears most commonly when discussing small-scale, rural or prehistoric societies (Hodges 1987; C. A. Smith 1976). It has, however, appeared less frequently in recent years, because the theoretical models of social evolution and cultural ecology have declined in popularity. In comparison, the use of the term community to refer to the inhabitants of an

archaeological site is still widespread. Again, it tends to be particularly common in studies dealing with rural settlements (e.g., Dyson 1992; Schwart and Falconer 1994), but it is also frequently used of archaeological sites in general (Deetz 1977; M. Johnson 1999, 99; Renfrew and Bahn 1991, 102; Verhoeven 1999). This assumed equivalence between the site and the community is not just theoretically unsound; it can also impose interpretive limitations. As already highlighted in chapter 1 of this book, this has been a particular problem for Anatolian archaeology. Here, the invention ofPage 26 → communities has been informed by the assumption of “betweenness,” and that assumption has led to the invention of communities. This continuing misuse of the term community can be largely attributed to the fact that there has been little critical discussion in archaeology of the community as a concept and no explicitly archaeological theorization of the term itself. There has, however, been some brief discussion in archaeology of the uncritical way that the idea of the community has been used. Roughly around the same time that the practice of community studies was coming under attack within anthropology (see chapter 2), the processual archaeologist Ruth Tringham began to criticize the use of the term in archaeology along similar grounds (1972). Specifically, Tring ham pointed out that trying to equate a geographical unit (the site or the settlement) with a social one (the community) was theoretically unsound. In this, Tringham echoed Stacey's damning 1969 critique of community studies. While archaeology may have been on the button when it came to critiquing bad theories, it was less quick off the mark when it came to developing new ones. The idea of community as a form of social identity did not take off in archaeology, despite its popularity as a concept in other social sciences. This was largely due to the outlook of processual archaeology in the middle and late twentieth century—it was argued that “soft” concepts like identity had no place in the “hard” science of archaeology (e.g., Hawkes 1954). The mere possibility of examining identity through material culture was summarily dismissed. Ruth Tringham went as far, in fact, as to suggest that it would be impossible to propose a connection between settlements and the groups of people who inhabited them: “On the basis of present evidence, it seems doubtful that any correlation between archaeological settlement and the sociological concept of community can be made” (1972, xxi; original emphasis). Tringham rightly rejected the unsound assumption that there was an automatic link between residential proximity and a social identity. However, this assumption was immediately replaced by another, just as unsound: that there was no link at all between the two or, if there was, that the link would be impossible to identify in the archaeological record. This new assumption, combined with the more general trend of the time that favored regional over site-scale analysis (Yaeger and Canuto 2000, 3), precluded the development of an explicit archaeology of communities for decades. Not until the turn of the twenty-first century did archaeologists begin to rediscover the concept and its myriad possibilities. Page 27 →

Toward an Archaeology of Communities Archaeology rediscovered the idea of the community only in the final years of the 1990s, over a decade after the concept's rehabilitation in social theory. This rediscovery itself was only made possible by a number of developments within the discipline, which allowed archaeologists to bypass the natural-deterministic and evolutionary-structural approaches that had dominated archaeological thinking about communities before. Most of these developments took place in the late 1980s and 1990s, as archaeology began to turn away from the strictly empirical approaches of earlier decades and started to engage more closely with questions of identity, subjectivity, and social meaning (Jones 1997; Meskell 2001; Shennan 1994; Trigger 1989, 444ff.). Connected to this and occurring around the same time, there was also a “spatial turn” in archaeology, with ideas of phenomenology, experience, and embodied practice beginning to be incorporated into archaeological interpretation (Gosden 1994; Hodder and Hutson 2003; J. Thomas 1996; Tilley 1994). The combination of these two developments now makes it possible for a new generation of archaeologists to start thinking about community as a form of social identity and about how community identity might be identified and analyzed through material culture. The first attempts to reevaluate the community as a focus for archaeological study harked back to both of the traditional archaeological approaches to communities. In 1997, Michael Kolb and Jason Snead argued that communities were a useful concept for archaeologists because they were a natural and therefore universal form of human organization common across rural and agricultural societies. As such, they suggested that looking at

communities would be a useful way of conducting cross-cultural comparisons. This perspective sees communities as naturally arising from residential proximity, in much the same way as Tönnies did, and it sees communities as unproblematically mapping onto sites, in much the same way as Childe did. At the same time, however, it also considers the community to be a particular form of structure, entailing specific forms of social hierarchy and organization that occur only in small-scale rural settlements and not larger urbanized ones (Kolb and Snead 1997, 613). The first steps toward an archaeology of communities did not, then, include a reevaluation of the use of the term in archaeology or an in-depth examination of the recent developments in social theory regarding communities. These first steps are nonetheless important because they stimulated interest and discussion. Page 28 → Not until the publication of Marcello Canuto and Jason Yaeger's 2000 edited volume, The Archaeology of Communities, did the study of communities return to the archaeological mainstream. Canuto and Yaeger's work argues the value of the community as a socially meaningful unit of analysis, belonging to the somewhat neglected medium scale between the household and the region. They recognize that residential proximity does not automatically lead to a psychological sense of community identity, but they argue instead that the two can be related through embodied experience and social practice. Building on the insights of phenomenology and Bourdieu's practice theory, they point out that the practical experience of living close together and regular interaction can potentially allow for a sense of corporate “us-ness.” This, they argue, can then be used as a basis on which to create a more conscious sense of community identity, through what Yaeger calls “practices of affiliation” (2000). Such practices both act to produce and are simultaneously the products of a conscious ideology of community togetherness. These practices can take many different forms, from explicit verbal statements of unity to the deployment of material culture styles. Therefore, while it recognizes that residential proximity does not lead directly to community identity, Canuto and Yaeger's formulation of community does see an important potential connection between the two. It casts community as a dynamic form of social identity, anchored to the idea of shared geographical space and common lived experience through the active use of shared social practices. Nonetheless, it still recognizes that a sense of community identity was neither a universal nor a consistent feature among the populations of nucleated settlements, and it readily acknowledges that community identity only arises in certain historical situations. Under this formulation, then, community identity is both socially constructed and flexible. It may also crystallize or dissipate depending on the wider historical circumstances. Yaeger and Canuto initially felt that this instability was a potential thorn in the side of a nascent archaeology of communities. However, this book argues that, far from being a failing in the archaeological study of community, instability is actually one of its most interesting facets. The points at which community identity becomes salient and the moments at which it ceases to be so are of crucial importance. When and why does the identity of the community as a group become more important than the various different forms of individual identity that intersect it, such as rank, gender, or age? In what situations do these other forms of identity become more important again, at the expense of communal group identity? Yaeger's own examination of communitiesPage 29 → from the classic Maya period illustrates the value of asking such questions (Yaeger 2000). He describes how the incorporation of local communities within a larger state led to fewer expressions of community identity, as local elites increasingly sought to emphasize links with the central state authorities and thus to stress their difference from the rest of the community. Interestingly, Pauketat's work on early Mississippian communities shows that a similar situation can also lead to the opposite response (2000). He points out that the assimilation of local communities into the wider Mississippian cultural sphere often prompted conscious assertion of a “politicized” community identity. The difference between these two cases speaks volumes about the different social dynamics at work within and between communities and offers valuable insights into both cultural and historical processes.

Developing the Archaeology of Communities Canuto and Yaeger's formulation of community has since been taken up by several other scholars, who have applied the concept of community as a socially constructed form of identity rooted in shared geographic space and experience (e.g., Gerritsen 2004; Knapp 2003; Peterson and Drennan 2005; Wernke 2007). However, while

Canuto and Yaeger's work has moved us much closer to an archaeology of communities, it is generally recognized that a comprehensive archaeological approach to communities is still lacking (Canuto and Yaeger 2001, 1; Gerritsen 2004, 151). The important steps forward taken by Canuto and Yaeger, then, are not excuses for complacency. For a start, there remains some confusion over just what constitutes a community, and the distinction between geographic and relational communities is frequently blurred even in the more recent archaeological literature on the subject. In addition, there is also considerable uncertainty over how a conscious sense of community identity can be inferred through the archaeological record. Finally, there are issues of advocacy and changing the understanding of communities in the archaeological mainstream. The nascent archaeology of communities has struggled to progress, at least partly because it has met with severe criticism. There is still some resistance to the “community” as a focus of research in archaeology, its detractors arguing that it is difficult and perhaps even impossible to identify social groups in the archaeological record that might have defined themselves as communities. Even within Canuto and Yaeger's edited volume itself, Page 30 →there was disagreement over the necessity of having a geographical aspect to community identity, and concerns were expressed that the Canuto-Yaeger formulation was no more than a disguised version of the old assumptions equating settlement with social identity (Isbell 2000). In particular, critics have argued that communities are “imagined” rather than naturally arising and that the inclusion of a spatial component in community identity falls too close to neoevolutionary determinism (Isbell 2000; Lekson 2002). The vehemence of these arguments masks a serious and ongoing problem for the “archaeology of communities”—that the use of site and community as equivalent and interchangeable terms is still prevalent, even among scholars who consciously cite Canuto and Yaeger's work (e.g., Düring 2006, Hodos 2006, 23; Keswani 2005; Westgate, Fisher, and Whitley 2007). Despite referencing Canuto and Yaeger, all these authors have continued to use the term communities to refer to the inhabitants of sites, without exploring the socially constructed nature of community identity. This uncritical use of the term may be due in part to the fact that the idea of the community is still undertheorized in archaeology as a whole. Not only do we lack a complete and coherent definition of the community in archaeologically relevant terms, but we also lack a practical methodology for the archaeological investigation of community. It is hoped that this book will contribute in some small way toward improving this situation. A second line of criticism that has been leveled at the developing archaeology of communities points out that many communities do not have a geographic or territorial component, and critics have cited many examples of diaspora communities and virtual communities where members rarely, if ever, meet. Proponents of the archaeology of communities themselves struggle with the occasional lack of geographical specificity in the communities they are studying (Goldstein 2000; Yaeger and Canuto 2000). This flexibility of terminology is deeply problematic for the current archaeology of communities. As remarked on at the start of this chapter, the term community is an open one and can be very broadly applied. As we understand them, communities can focus on shared professional, political, or religious interests. They can span nations and continents. Significantly, they can interact on either a remote or a face-to-face basis, and sometimes they can even exist at the level of the individual imagination and do not necessarily require any social interaction at all. Given that the term community can be applied to almost any form of group identity, how can archaeology ever hope to investigate communities? This is the position argued by some critics, who doubt that the term can ever be rendered practically useful for archaeologists (DeMarrais 2002; Lekson 2002). Page 31 → The central problem underlying both of these criticisms is one of definition. No clear and archaeologically relevant definition of the community has yet been agreed on, and as a result, the word tends to be used as a general term to refer to anything from the site to the identity group. However, as we have seen in chapter 2 of this book, other academic disciplines have already developed tighter definitions of the community. Drawing from these insights, it is possible to assuage the concerns of the critics and move forward to develop a coherent and theoretically sensitive archaeology of communities. One criticism derives from the current broad usage of the term community to apply equally to any kind of collective identity, from ethnic-based identities to political ones. The distinction between geographic and relational communities, which is now common throughout the social sciences, should come into play here (see chapter 2). Groups that have a sense of common identity based on a perception of

shared interests should be termed relational communities. Groups that live within a given locality and that also have a sense of common identity should be termed geographic communities. Most archaeological discussions about communities fall into this latter category, given that archaeology is, by nature, a spatial discipline (see the next section of the present chapter). The other criticism stems from the lack of a distinction between the population of a nucleated settlement and a social group that shares a conscious sense of collective identity related to that coresidence. As discussed in chapter 2, communities cannot be equated with settlements or neighborhoods. Instead, they should be understood as identity-bearing social groups that are focused in a specific geographic locale. Given a full and detailed theorization of communities themselves, then, the critics of the emerging archaeology of communities can be answered. A theoretically sound archaeology of communities is therefore possible. As seen in chapter 2, theorizing the community is no easy task, and a single comprehensive definition remains elusive. Nonetheless, the understanding of the community arrived at here should answer the main arguments against the current archaeology of communities. It should also allow for more progress. This book hopes to build on the foundations already laid for an archaeology of communities, developing the existing ideas and critiques into a coherent theoretical and methodological framework for approaching communities in archaeology. This framework is not expected to be unassailably robust or foolproof—it is likely to require revisions, additions, and partial demolition in the future. However, the goal of this book is to open up debate about the archaeology of communities. It aims to present a concise and comprehensive structure for approaching the subject,Page 32 → but it should be emphasized that this is only a working structure—a point from which to start. It is time to put an archaeology of communities firmly back on the agenda.

Geographic Communities and Archaeology In the remainder of this book, the word community will be used specifically to designate the geographical community—an identity-bearing social group whose conscious sense of collective belonging is rooted in the experience of residential proximity and shared space. Like any other form of group identity, the geographic community is built on and around some vital criteria: namely, perception of commonality between members and distinction from nonmembers (A. P. Cohen 1985, 115ff.). Again, like in any other type of group identity, this perception of commonality must be created and maintained through social practice (Bourdieu 1977). What is particular to the geographic community, however, is the form of the shared experience in question. Just as the common experience of a religious group identity pertains to shared forms of spirituality and cult practice and as that of the ethnic group relates to shared putative descent, the shared experience of the geographic community is that of communal space and geographical location. It is the spatial and emplaced nature of the geographical community that makes it particularly appropriate for study in archaeology. The archaeological record is inherently spatial in nature. Archaeological material is usually collected spatially, whether through excavation or survey, and it is also interpreted within a spatial context (Gamble 2001, 139ff.; Greene 2002, 50-51; Hodder and Orton 1976; Renfrew and Bahn 1991, 75ff.). Archaeology is therefore particularly well placed to identify the traces left by a spatially focused sense of identity. I have chosen to describe the geographic community as a “spatially focused” identity for specific reasons. When approaching geographic communities, there are two ways of looking at their emplaced nature—in terms of either spatial focus or territorial extent. Spatial focus emphasizes the central places of the community, places where the community concentrates its activities and that are nodes for interaction. Territorial extent, by contrast, emphasizes the spatial peripheries of the community, the boundaries of the community where insiders are demarcated from outsiders. Both spatial focus and territorial extent are necessarily important in the construction of a geographic community identity, but for the purposes ofPage 33 → this book, I shall argue that it is the former that is the most crucial. After all, a sense of geographic community is largely dependent on the existence and practical use of central community spaces. These central spaces are the key locations where community identity is created and around which it is focused. Central spaces do not need to be physically central to the community. Rather, they are central in a conceptual sense, meaning that social life, interaction, and everyday activity are focused around them. Central

spaces can therefore be spatially peripheral at the same time as being socially central. As can be shown by numerous modern examples, such spaces facilitate public activities and the interaction between community members necessary for the creation and maintenance of group identity (Karp 1997; Schorr 1997). From the first urban geographies of the Chicago school, scholars have identified central public spaces as both social and spatial nodes around which community life revolves (Lynch 1960). More recently and in more specific circumstances, town planners redesigning the layouts of American suburbs have highlighted the importance of public spaces such as parks and civic centers, pointing out that such communal spaces act as focuses of community pride and sense of place (Langon 1994). Similarly, village social clubs or churches are often important focuses for communal activity and venues where the community can get together to enact its sense of collective identity (Mitchell 2005). Archaeology is especially well suited to the investigation of geographic community identities because it is well equipped to study such central places and the activities that took place within them. Central places are often recognizable in the archaeological record by their architectural features, which may include large open areas where substantial numbers of people can gather and architectural ornamentation or monumentality. In cases where these spaces are physically as well as socially central, it may also be possible to recognize them from the spatial organization of the settlement around them. It is frequently also possible to discern the traces of the communal activities that potentially allowed for the creation of a corporate identity. These communal activities can often be inferred from the quantity and scale of evidence found—large deposits of dining wares as opposed to a small service set, for example. Identifying public and communal activity in these ways has a long history in archaeology. At Rome, for example, a conscious sense of place and communality was encouraged in annual public festivals such as the Lupercalia, which celebrated the city's foundation. Such festivals were staged in central spaces and involved specialized equipment and objects (Rykwert 1976). Page 34 → Archaeology is less well equipped to investigate the territorial extent of a community than its central focus. Even in cases where there are clear archaeological remains indicating a spatial boundary, historical evidence warns us to remain wary. The relationship between the ancient Greek polis and its agricultural hinterland neatly illustrates the problems with equating a nucleated settlement with a single geographical community (Osborne 1987). Similarly, villages in rural England are often considered as incorporating outlying farmhouses beyond the immediate settlement area (Mewett 1982). Conversely, within larger settlements such as cities and large towns, a geographical community can often be linked to a particular neighborhood or quarter, so that several communities can exist within a single nucleated settlement (Glynn 1986). Thus the limits of nucleated settlement do not always imply the spatial limits of a geographic community. Several different geographic communities can coexist within a single nucleated settlement, and geographic communities can also extend beyond the boundaries of the settlement. It is extremely difficult, then, to determine territorial extent through the archaeological record alone. Fortunately, territorial extent is less important for the creation of community spirit than central focal spaces. First, while central spaces are often in fixed locations, territorial boundaries can often be more fluid. Second, it should be recognized that sense of community is created less through mere shared experience of moving around a territory than through conscious social practices. These practices and the creation and negotiation of group identity take place at central focus places. Ethnographic parallels show that even in communities that place great emphasis on territorial boundaries, such as arctic hunters, the creation of community identity still focuses on certain specific places that are considered to be socially and symbolically central (Brody 1981). This discussion has already begun to move from the intricacies of what defines a geographic community to the dynamics of community identity itself. How exactly does a sense of geographic community identity coalesce? How does the shared experience of space and location contribute to the formation and continual re-formation of community identity? How do central spaces get utilized to promote a sense of collective togetherness? What kinds of social practice are involved? These questions lead us to consider the double nature of identity. Every type of identity has two components; a sense of self and a sense of other, a feeling for what constitutes “us” and what constitutes “them.” The next chapter discusses these components of geographic community identity.

Page 35 → CHAPTER FOUR

From Community to Community Identity The previous two chapters of this book have discussed the concept of the community as it has been approached, first, in disciplines such as geography, sociology, and anthropology and, second, in archaeology. In these chapters, it has been established that the community is primarily a form of identity—a mental construct rather than a natural or structural phenomenon. Arriving at a detailed but theoretically sensitive understanding of the community, however, is only the first step in developing a new archaeological approach to communities. The next step is grasping the dynamics of community identity—examining how the mental construct of community is realized through social practice and how a sense of community belonging is created, affirmed, and contested. To do this, we must turn from thinking about the community to thinking about community identity.

Group Identity and Creating a Sense of “Us” Sharing space on a regular basis is a powerful phenomenological experience, which informs an individual's conscious perception and unconscious cultural expectations. Equally, however, the inverse is also true—the individual's perceptions and cultural expectations have an impact on space itself and other people's experience of it. The experience of the environment therefore both shapes and is shaped by human perception. This dynamic relationship between mind, body, and space has been explored in a number of different philosophical and cognitive theories, including the concept of embodiment (Rohrer 2007), the theory of habitus (Bourdieu 1977), and the idea of phenomenological Being (Heidegger Page 36 →1927). The insights gained from these theories suggest that the shared phenomenological experience of coresidence can potentially provide the basis for a conscious sense of commonality. In addition to this, the practical realities of sharing space mean that coresidents often share some social experiences as well as psychological or phenomenological ones. It is highly likely that people who live in close proximity have similar experiences of social practices and activities that occur within their shared space. These commonalities can thus also provide a foundation for the realization of a community identity. Social practice (Bourdieu 1977), interaction (Barth 1969; Giddens 1984), and encounter (Buber 1947, 1958) have been identified as important elements in identity formation—once again, both informing and being informed by human perception. Over the last two decades, archaeology has incorporated many of these ideas about phenomenology and social practice into its interpretive frameworks, developing these ideas for a specifically archaeological context (e.g., Hodder and Hutson 2003; J. Thomas 1996). In their pioneering discussions of community consciousness and archaeology, Canuto and Yaeger have built on this existing body of archaeological theory (2000). They point out that a sense of community is rooted in social experience and social practice and that it is through these shared experiences and practices that identity is continually created and re-created. However, while Canuto and Yaeger's model focuses closely on the idea of social practice, it does not fully consider the emplaced nature of this social practice. In contrast, this book argues that the geographic focus of shared experience is the key distinguishing feature of community identity. What separates community identity from other types of group identity is its sense of location. The rooting of social practice in common space is vitally important in understanding the dynamics of community identity (Bell and Valentine 1997, 94; Sack 1992, 188). It should be recognized that the social experiences shared by a community are firmly located in geographic space and that, consequently, this space is likely to take on some social significance (Relph 1993). Although individuals can have very different experiences of the same environment (J. Thomas 1996, 83ft), the fact that emplacement and location are the common foundations of this experience should not be glossed over. In short, the phenomenological experience of living together makes it possible for a sense of community identity to arise. Locality ensures that coresidents will have some shared phenomenological experience, as well as common elements of social habitus. Taken together, this forms a foundation Page 37 →from which a conscious sense of community identity can potentially be crystallized.

While a conscious sense of a community identity can possibly be drawn from the experience of coresidence, it is by no means certain that this will necessarily occur. For this conscious sense of a collective identity to emerge, the perception of that identity must be deliberately constructed through social practice. A sense of community, therefore, does not arise automatically from commonalities—it surfaces only when commonalities are consciously acknowledged and deliberately celebrated. Commonalities can be emphasized in several different ways. Explicit statements of unity can be made through verbal or literary means or through artistic representation. There is already a substantial body of literature that uses these modes of evidence to discuss community identity in historical (e.g., Simons 2001) or modern (e.g., Amit 2002) societies. However, literary or ethnographic source material is not often available for prehistoric and protohistoric societies. Archaeologists rely instead on material culture to explore ancient identities. Fortunately, a sense of community unity can also be expressed through performance—through praxis or social activities. Such activities are the experienced enactments of community or group identity, and the traces of these are sometimes discernible in the archaeological record (Mac Sweeney 2009). Enactments of community are social practices and activities through which a community consciously emphasizes its perceived togetherness and promotes the ideology of group solidarity. Such practices are referred to as “practices of affiliation” by Yaeger (2000) and as “affiliation dramas” by Strathern and Stewart (2000). I have decided not to use either of these terms in this book, as the concept of affiliation implies a pre-existing collectivity, which individuals become affiliated to. In contrast, I would argue that communities themselves are created through these social practices. Indeed, enactments of community are necessary because social cohesion is never automatic—other forms of social identity, such as gender, status, or age, divide communities (Diaz-Andreu et al. 2005; Knapp and Van Dommelen 2008; Meskell 2006). Whereas community identity plays down the differences between individuals, glossing over them with an ideology of collective sameness, these other forms of identity focus on specific points of difference and emphasize social differentiation. Over the last few decades, archaeology has focused much of its attention on intracommunity dynamics, considering the way social divisions are made and distinctions are maintained (Diaz-Andreu et al. 2005; Gamble 2007; Page 38 →Insoll 2006). In particular, many archaeologists have focused their investigations on concepts of agency, personhood, and the individual in society (e.g., Dobres and Robb 2000; Dornan 2002; Fowler 2004; Knapp and van Dommelen 2008). Such work has been very important, leading to a greater understanding of how societies can be structured and restructured and the ways in which ideologies can be interpreted, reinterpreted, and subverted by different groups and individuals. However, this trend has meant that relatively little research has been done in recent years on social cohesion and collective group identities (Mac Sweeney 2009; Osborne 2007). A situation remains in archaeology where the existence of group identities tends to be assumed rather than actively investigated. As already discussed in chapter 3, nowhere is this more common than in the tendency to ascribe a sense of community identity to archaeological sites; and as highlighted in the introduction, this is especially common in the case of western Anatolia. The study of social cohesion does not deny the existence of differentiation or of individual identities. Instead, it offers complementary insights into social dynamics. Group identities, like individual identities, are flexible social constructs, which only become salient in specific historical situations for a specific set of social reasons. There will be certain moments in time at which a community identity becomes important and moments when it subsequently ceases to be so. These moments are of crucial significance. When and why does the idea of the community as a collective come to the fore over the various forms of identity that intersect it, such as rank, gender, or age? What kind of circumstances lead to situations in which the ideology of communal togetherness trumps that of internal social division? In what situations do individual identities and the desire for social differentiation return to being more salient than communal group identity? These questions, fundamental to understanding group dynamics, demonstrate that a study of community identity complements and does not contradict studies of social differentiation. Community identity is not, therefore, the same thing as social uniformity and equality. Rather, it is a particular form of social ideology that emphasizes sameness and cohesion rather than difference and differentiation. There will always be some things that members of a community share and some things that they do not. Most forms of identity emphasize what is different, playing up these differences to create a sense of social differentiation. Community identity does the opposite. It emphasizes what is shared, so that social differences are played down and temporarily pushed aside. It focuses on creating, consciously and

artificially, a sense of “us.” In the case of geographic communities specifically, Page 39 →the foundation of what is shared lies in the common experience of place and locality.

Sociospatial Distinctions and Creating a Sense of “Them” Any sense of identity is concerned with both sameness and difference (Barth 1969; A. P. Cohen 2002). A sense of collective us must be matched by a sense of “them,” or the external Other. Not only must there be a conscious sense of us, but there must also be a conscious sense of them. These necessary opposites can only exist in relation to each other and are constructed in opposition to each other. To investigate community identity therefore, it is vital not just to establish the evidence for a conscious sense of us but also to consider the evidence for them as well. But while the central focus point of the community is important for a sense of us, the boundaries of the community are essential for a sense of them. At these boundaries, the distinction between insiders and outsiders is made, and the criteria for community membership are established. In the case of geographic communities, we might assume that the primary criterion for inclusion and exclusion is necessarily geographic. Being physically beyond the bounds of the community is an obvious criterion against which to define nonmembers. However, geography is not the sole or sometimes even the most important criterion for community membership. As already discussed in chapter 3, the spatial boundaries of a community are often blurred, and it is not always possible to establish a clear geographical limit to the community. In addition, it is also possible for people to be considered as members of the community while being at some geographical remove, as may often be the case with emigrants or travelers. Finally, it is also possible for individuals within the spatial boundaries of the community to be excluded from the community socially, such as the single mothers who face social exclusion in some modern American communities (Brodsky 1996). It is clear that geographic factors alone do not determine exclusion from the community. It has already been established in the preceding chapters that community identity has both spatial and social aspects, and we should therefore assume that there will be both geographic and social criteria at play in defining the membership of any community. Indeed, one only needs to consider modern and historical examples to find that additional social meanings are often layered onto a community identity, making community membership based around other things in common as well as geography. Page 40 →For example, many communities through history have shared occupational commonalities as well as geography, including the mining communities in northern England during the early to middle twentieth century (Bell and Newby 1971, 166ff.; Dawson 2002) and communities focused on particular agricultural niches (Gray 2002). Communities within Roman military forts were not simply predicated on residence within the space of the fort and surrounding town but were also closely concerned with the military profession as well (Gardner 2007). Similarly, there are many examples of communities that share not just a locality but also a religious focus, such the Congolese villages that organize themselves around a sacred miyoomb tree (De Boech 1998). While community belonging may initially be predicated by the acknowledgment of coresidence, additional social factors come into play more often than not. The social factors at work can be complex and can embrace issues as broad as ethnicity, sexuality, religion, profession, economic standing, linguistic capability, political preference, or sporting activity. The net social meaning of the community's identity, made up of the combined social factors overlaid on top of the initial geographic identity, is referred to in this book as the social rationale of community identity. The social rationale of a community identity is not only complex in itself but also changeable. This fluidity is evident in the shift in the community identity of the Notting Hill area of London, which in the late 1950s and early 1960s was characterized by social features including a high Afro-Caribbean immigrant population and poverty. Ethnicity and low socioeconomic status were therefore the defining social elements of community identity at this time (A. P. Cohen 1993). Something of the atmosphere of the time was captured by the Southam Street project by the photographer Robert Mayne (Mayne 1993) and in the work of the Trinidadian filmmaker Horace Ové (Givanni 1996). By the late 1990s, however, the area was characterized instead by a young, wealthy, and predominantly white population, and wealth and high socioeconomic status took over as the main social characteristics of the community (York 2006). There have been varying reactions to this change; for some, it is the subject of cynicism and satire (R. Johnson 2006), while it is celebrated by others, most notably in the 1999 film Notting Hill, directed by Roger Michell. The social rationale of community identity is therefore both complex and

changing. It is specific to each individual community and to each historical time. As such, it is impossible to generalize about what social factors might be relevant in a given community, and these can only be identified on a case-by-case basis. Page 41 → A community's social rationale may therefore be complex, changing, and historically situated. But it should now be evident that the social rationale is just as central to community identity as the geographic focus. While the geographic focus may be where a sense of collective us is constructed, a sense of them crystallizes at the social boundaries of the community. The social rationale of community identity is therefore vital in creating a conscious sense of them. This chapter has focused on the dynamics of identity in a geographic community, building on the understanding of different types of community reached in earlier chapters, to consider process rather than definition. The geographic community is a form of group identity that is fundamentally rooted in the shared experience of place. This spatial aspect distinguishes geographic communities from other types of relational community and makes them a suitable subject of study for archaeology. This chapter has discussed how emplacement and common location can contribute to a mental construct of identity, considering the processes by which this contribution can happen. It has also discussed how the mental construct of this identity can be realized through social practice, to create both a sense of us and a sense of them. It was established that a sense of us is created by emphasizing commonality and suppressing social differences and that this is achieved through enactments of community. In comparison, a sense of them is constructed by building up a social rationale of community identity that imputes certain social characteristics to insiders and outsiders, adding social meaning to the geographic basis of identity. We have therefore explored the dynamics of geographical community identity and considered how it operates. But how can this be investigated in archaeology? It is now time to consider how material culture is used in enactments of community and how traces of this may be perceptible in the archaeological record.

Page 42 → CHAPTER FIVE

Community Identity and Material Culture Using Material Culture The relationship between culture and identity has been the subject of much academic discussion, with the central issue being how culture, in both its material and nonmaterial forms, relates to identity as a cognitive construct. Identity is now widely seen as a social dialogue formed in the interaction between internal psychology and external lived experience. The active potential of human agency is seen as working around a social and environmental framework, a relationship that has been explored by philosophers and social theorists from Heidegger to Weber, Bourdieu to Giddens (Turner 2000). Community identity is no exception to this. The psychological sense of community is shaped by the experience of living in a community, but at the same time, it also creates this very experience. The formation of community identity, then, is analogous to the formation of other types of social identity and takes place in the third space between psychology and experience (Soja 1996). It is an ongoing and discursive process, in which psychology and lived experience are linked in a continual and mutually dependent feedback loop. The psychological side of this loop is often investigated through written or spoken testimony, both of which offer explicit discussions of identity and explicit accounts of its social meaning. It must be borne in mind, however, that such accounts are not straightforward statements that passively reflect the individual psychologies underlying identity. Rather, they are consciously constructed representations of identity operating within a social context. Both texts and ethnographic sources can potentially distort and construct social realities as well as represent them; equally, they can be manipulated by social actors in the service of social Page 43 →strategies. Both written and oral testimonies therefore actively construct identity, rather than simply reflecting it (Bhabha 1990; Cameron 1989; White 1973). While these forms of evidence may be well placed to shed light on the psychological half of the identity feedback loop, material culture plays a comparable role for the experiential half, offering perspectives onto lived cultural experience. The lived experience of an identity is comprised of its practical enactment in various social relationships, enactments that are negotiated in the physical world and leave discernible material traces. This concept is central to current theoretical archaeology (Gamble 2001, 196–202; Hodder and Hutson 2003, 106ff.). But material culture is more than the passive traces left by identity interactions, as has often been pointed out. Like text, it can be actively used to construct identity and can be manipulated with reference to specific social contexts (Hodder 1982, 1989; Tilley 1989, 1990, 1992). Material culture therefore provides evidence for the lived experience of identity, and oral/literary testimony for explicit rationalizations of it. Texts/ethnography and material culture are therefore complementary forms of evidence representing different sides of the identity feedback loop, and investigations of identity would ideally have recourse to both. This is not always possible, however, and many archaeological studies of identity must rely on material culture alone. But while material culture is theoretically just as indicative of identity as text, realizing its full potential has, in practice, proved difficult. The study of social identities through material culture raises certain challenges. First, it is not always certain which material traces of cultural experience relate specifically to identity formation and which do not. Second, there is ambiguity surrounding the social meaning of identities even when cultural experiences can be distinguished as relevant to identity formation. Two distinct analytical processes are therefore necessary: (1) the identification of evidence for identity and (2) the interpretation of that evidence. These processes should both be applied in the study of community identity in archaeology. First, material culture should be examined to determine whether or not there is evidence for a conscious sense of community identity. This evidence will most likely take the form of the shared social practices that function as enactments of community. Only once this evidence for a conscious sense of community is identified can its social meaning be interpreted. There is a tendency in archaeology to assume the centrality of ethnicity to all group identities

(Broodbank 2004; Mac Sweeney 2009; Shanks and Tilley 1987). However, this tendency must be Page 44 →resisted; instead, the social rationale of a community identity must be inferred from the wider social and historical context. This chapter proposes a methodology for carrying out these two distinct archaeological processes in the archaeological investigation of community identity. This methodology builds on existing theories of visual recognition, cultural experience, and stylistic communication, as well as the ideas about the relationship between culture and identity previously outlined. It will consider what types of material traces might be left in the archaeological record by the performance of community identity. In the next section, it will consider the possible traces left by a sense of collective us; in the following section, it will review the potential archaeological evidence for a sense of external them. Finally, in the concluding section, it will explore how the various forms of evidence for community identity can be placed within the wider social and historical context to infer the related social rationale.

Inferring the “Us”: Enactments of Community Identity is not a passive social experience but an actively created and ascribed social construct. Group identities are constructed by promoting a sense of unity and togetherness (Mac Sweeney 2009), a process described in the “practices of affiliation” referred to by Yaeger (2000). A similar social process is also identified in the “affiliation dramas” staged in the highland communities in New Guinea (Strathern and Stewart 2000). A conscious sense of community identity does not arise automatically or naturally from mere residential proximity. The experiences of sharing space and location provide a basis on which community identity can be built, but they are not sufficient in themselves to ensure that community identity will crystallize. To infer the existence of a conscious community identity, there must be signs that people chose to celebrate a sense of collective “us-ness.” The concept of “enactments of community” was introduced in the previous chapter. These are social practices that encourage a sense of social cohesion by bringing the community together, highlighting commonalities and glossing over internal divisions. Within archaeological contexts, therefore, the existence of community identity can only be inferred if evidence can be found for enactments of community. Finding such evidence, however, is far from straightforward. These enactments of community can take several forms, and there is no fixed checklistPage 45 → against which they can be identified. However, they do have some recurring features. First, enactments of community often involve the participation of a significant proportion of the community. Even if many of the participants are only engaged as spectators, the experience of being together and sharing in a group activity still acts to thereby foster a sense of collectivity. This is borne out by historical and ethnographic examples demonstrating that social cohesion and group identities can be promoted by the staging of large-scale, public events (Bell and Newby 1971, 30). Within the rituals of twentieth-century Balinese temples, for example, the communal cooking and sharing of rice is effected in such a way as to encourage collective identity, explicitly acting to obscure status and ethnic distinctions (Ottino 1998, 116–19). Similarly, within the provinces of the Inca Empire, hosting large-scale ceremonial feasts was an important means by which the central state sought to promote its ideology of integration and unity throughout the empire (DeMarrais, Castillo, and Earle 1996, 27–28). Signaling theory also makes this connection between public events and identity formation (Bliege Bird and Alden Smith 2005). It is pointed out as a simple principle that for a social statement to have maximum effect, it should be made in an appropriately large social venue and in front of an appropriately large audience. Or, put more simply, when making a statement about identity, it clearly helps to have someone to make it to. Beyond their tendency toward wide participation and public engagement, enactments of community are also likely to be practices invested with social significance and symbolism (A. P. Cohen 1985, 50). One arena that may be relevant here is religion. Durkheim suggests that, at a basic level, religion often acts to encourage solidarity and social cohesion (Durkheim 1912; Morris 1987). One example of this is Neolithic Malta, where unique forms of ritual practice and architecture were developed in the fourth millennium BC. The distinctiveness of the temple tradition promoted both a sense of social unity between those who used them and a sense of difference in the face of outsiders (Robb 2001). But even nonreligious ritualized activities are likely to be at the heart of socially significant events and practices where identities are formed (Leach 1968). This is evident from ethnographic

studies, such as the research on the San in the Kalahari (Wiessner 1984). Social identities here are articulated most potently in public contexts invested with social significance, usually during ritual or ceremonial events, or at events “when one's heart soars” (Wiessner 1984, 200). Page 46 → The example of the San, however, highlights the potential ambiguity of such social practices. The same events where community identity can be constructed may also be venues for the negotiation of individual and subgroup identities. Large-scale public feasting, for example, can be a forum for the creation and expression of individual rank and status, as it seems to have been in Iron Age France (Dietler 1990, 1999). Equally, however, it can also be a focus for the promotion of a community identity (Bell and Valentine 1997, 106–7). For such an event to function as an enactment of community rather than emphasizing individual intracommunity identities, it must promote homogeneity rather than heterogeneity, uniformity rather than distinction, sameness rather than difference. This does not mean that communities were actually homogeneous or uniform, however. Sameness and community cohesion is an ideology—one form of ideology among many that can potentially act in a social group. Enactments of community are therefore social practices that promote this particular ideology of sameness, emphasizing it at the expense of other ideologies such as social ranking or gender roles. Within the same community, different ideologies may be pushed to the fore at different times and in different situations. Only if there is evidence of practices that promote the ideology of communal unity—that is, only if there are enactments of community—is it possible for us to infer the existence of community identity. In addition to broad participation and ritualized activities, enactments of community are therefore likely to conspicuously promote an ideology of unity or sameness. This can probably be traced in the archaeological record. Wide, open spaces are the most probable settings for enactments of community, as these permit large numbers of people to gather together. The opposite of this, enclosed space, is much less conducive to such communal gatherings, as it limits the number of participants. Similarly, structured or subdivided spaces also work against the concept of social cohesion, as they enshrine social divisions within the environment. There are therefore different types of social ideologies at play in different types of ritual space, and the collective group ideology of community identity can only be enacted in spaces that are appropriate to this ideology (Wright 1994, 61). Let us take eating lunch in a school dining hall as an example. An instance where there is a large, open, and undifferentiated space and in which there are no restrictions on how students are seated might encourage communality and cohesion between students. If, however, tables within the hall are reserved for different yeargroups, with students seated by their year-group at specifically allocated tables, such differentiated space may reinforce age divisions within the student body. Page 47 → Aside from the space in which enactments of community take place, the form of the material culture used in such practices must also be appropriate to the ideology of unity. In general, enactments of community are most likely to make use of large quantities of very similar objects. The emphasis on sameness requires that the participants undergo a similar experience, and the physical objects involved in this experience are an important part of this. If a social practice entails differentiation in the use of objects—either in cases where participants are engaged with a different quantity or quality of object or in cases where only some among the participants are engaged with objects at all—then it is the differences within the community, rather than the sameness across the community, that will be reinforced. To return to the example of school lunches, if students all eat their lunch from similar plates, using similar dining utensils, this will encourage a sense of equality between them. If, however, students in different year-groups are provided with a different quality of plate, this will heighten a sense of distinction. This can work at the level of horizontal distinction, as a group given glass plates may feel distinct from a group given fine ceramic plates. However, this can also work at the level of vertical hierarchy; a group given ceramic plates may feel superior to a group given plastic or paper plates. Similarly, if different year-groups are provided with different quantities of utensils, distinctions are also emphasized. If older children, for example, are given different cutlery for their main courses and desserts, this can highlight a distinction between them and younger children who are provided with one set of cutlery for all courses.

Both the space in which a social practice takes place and the objects that are employed in that practice are clues as to the nature of that practice. Enactments of community can therefore be identified by their inclusive and homogenizing nature. They are likely to take place in large, undifferentiated spaces and to involve the large-scale and uniform provision of artifacts. The example of school lunches may be particularly instructive, given that group dining and commensality is a social practice that can often be identified in the archaeological record (Hayden 2001, 40–41). It should be noted, however, that none of the spatial configurations or utensil provisions discussed in the school example mean that divisions within the student body do not actually exist. Even the most communal of school lunches does not deny the relevance of individual identities or the practical existence of yeargroups. Nonetheless, within the social practice of eating lunch, there is potential for two different types of ideology to be promoted. Differences can be played up to articulate a heightened sense Page 48 →of difference or played down to articulate a sense of unity and shared belonging. This section has dealt with features that encourage a sense of social cohesion within the community. However, the social construction of community identity involves twin processes: encouraging a sense of social cohesion and group togetherness within the community and also articulating a sense of difference from those beyond the community.

Inferring the “Them”: Social Rationale For a sense of community identity to crystallize, it is important not only that there is a sense of cohesion and shared identity within the community but also that there is a sense of Otherness and distinction from other communities. As already discussed in chapter 4, this sense of Otherness necessarily involves both social and spatial distinction—they must be distinguishable using both geographical and social criteria. The argument has already been made that of the two, the social criteria of belonging—the social rationale of community identity—most effectively determines the definition of them. Once evidence for community spirit has been identified in the form of enactments of community, an archaeological investigation of community identity must therefore turn to the social rationale of that community identity and the definition of the Other. However, it is not always straightforward to identify the construction of the concept of them or the Other using only material culture. By contrast, it can be relatively simple to recognize when a group is depicted as Other in explicit verbal statements. Verbal representations and literary discourse are often used to create a sense of Otherness or social distance, and this has been widely discussed in relation to a range of historical periods and a variety of literary traditions (e.g., E. Hall 1989; Hartog 1988; Said 1978). Groups can also be cast as Others using artistic representations, and this method of Othering has also been widely studied (e.g., B. Cohen 2000; Roberts 2009). In comparison, the use of material culture to cast certain groups as Other is less well understood. However, material culture is central to the way external groups are represented in social performance, and this representation can also cast these groups as Others. The performative representation of external groups consists of how these groups are portrayed through social practice and in embodied experience. The material construction of a concept of them and of the social rationale of community identity is closely dependent on treatment of physical Page 49 →objects and visual styles that recall external groups. We know from contemporary experience that objects can evoke associations with certain people, places, or ideologies. These objects can therefore act as symbols that represent the people, place, or ideology they recall (Appadurai 1988, 38; Knappett 2005, 119). The first stage of investigating social rationale through material culture therefore involves identifying items in the archaeological assemblage that recalled or evoked a sense of certain external groups beyond the physical bounds of the community and therefore could act as symbols of these external groups. However, it is not the mere presence of such symbols but the way these symbols were viewed, used, and treated that created the idea of the Other. In particular, the way that these symbolic objects were used in social situations both shaped and was shaped by perceptions of the things they represent. This process—the symbolic use of styles and objects to create an image of the things they symbolize—is termed performative representation, and it is through such representation that some external groups can be cast as Others by a community. Not all external groups, however, will necessarily be cast as them at the same time. Styles and objects that are symbolic of some groups might be embraced and integrated within the social life of the community, while styles and objects that are symbolic of other groups might be held at arm's length. This choice of who to accept and who to reject through the

symbolic treatment of material culture is significant of the community's social rationale. The differential treatment of various external groups places the focus firmly on the social criteria for community membership—after all, all of these groups are spatially distanced. However, some of these groups will be treated as if they are socially distanced as well. This selection of who to cast as them offers insights into the social rationale of a given community identity. A community's representations of external groups, whether literary, artistic, or performative, are distinct from the communities' actual or practical engagement with these groups. A community can happily engage in close trade relations with an external group, for example, while simultaneously representing them in a damningly negative way that casts them as Other. In classical Athens, for example, the violent anti-Persian rhetoric of the literary sources contrasts with the regular incorporation of Persian-style objects into everyday Athenian life (Miller 1997). Therefore, the representation of the Persians as distinctly Other did not preclude the use of Persian artistic styles and objects in Athens. At the other end of the scale, it is equally possible for a community to have little or no practical contact with an external group but, at the same time, to represent that Page 50 →group as being closely and intimately related. During the 1990s, for example, Albanian Kosovars often represented themselves as having close political links to the state of Albania, despite the decision of the Berisha administration to remain neutral over Kosovo's political status (Judah 2002, 96ff.). The performative representation of an external group is not, therefore, the same thing as the mere presence of an object or style that recalls that group. Rather, it consists of the way such objects are used and treated. What tells us who was cast as them is not the traces of external contact but the traces of how those contacts were represented and rationalized in the wider social context. The social rationale of community identity is therefore distinct from the practicalities of the community's actual engagement with groups beyond its boundaries—it is crucially concerned, instead, with the perception of these groups. Recalling External Associations The starting point for an archaeological investigation of social rationale is the identification of objects and visual styles that would have evoked or recalled associations with groups beyond the physical limits of the community in question. But what kind of objects recall external associations? What artifacts prompt their viewers to think of other places, other people, other communities? The most obvious answer to these questions is objects that are actually from these other places, which were made by other people, which are recognizably characteristic of other communities. However, this answer itself needs some careful unpacking. Archaeology has long been concerned with the identification of contacts between sites (Renfrew and Bahn 1991, 335ff.). Contact between sites is usually established by the identification of artifacts or cultural features from one site at another, and a complex and varied archaeological literature has been built up around how such external features may be recognized. While scientific provenancing studies are of great utility in recognizing such features, much of this work in archaeology is still based on the visual analysis of material culture styles. This analysis is usually carried out with reference to detailed typologies of formal and decorative elements. Such classification schemes are a standard feature of archaeological practice and are vital for dating and chronology as well as for understanding the spatial distribution of different cultural forms (Greene 2002, 141–47; Renfrew and Bahn 1991, 111–17). These typologies define material culture styles by the presence and absence of predetermined stylistic elements, elements often initially selected with the goals of quantificationPage 51 → and “scientific” objectivity in mind. This form-led approach stems partly from archaeology's disciplinary roots in antiquarianism (Trigger 1989, 80ff.) but is also partly due to the high value placed on quantification and objectivity in the New Archaeology of the 1960s and 1970s (see chapter 3). Studies of similarity and difference in material culture styles still tend to involve the compilation of a fixed list of stylistic elements and a statistical reckoning of the presence of these different elements. Methods such as these are both necessary and extremely helpful as tools for thinking about stylistic variation, and studies using these methods have provided many useful insights on the cultural dynamics and interaction in past societies (e.g., Glatz 2006, 2009; Whitelaw and Morgan 1991). The rigidity of stylistic typologies, however, does often have to be relaxed—and a certain elasticity often has to be factored in—when undertaking typological analysis aimed at cross-cultural, rather than chronological, investigation. The identification of “imitation” Mycenaean vessels in the central Mediterranean, for example,

allows for the fact that local imitations frequently include substantial deviations from the canonical Aegean forms. While these vessels generally fit the overall Mycenaean pattern, local innovations in style and indigenous preferences for shapes and sizes (as well as differences in production technique and clay types) distinguish them from the imported vessels. It is likely, however, that the communities using these vessels would have considered them to have been of a Mycenaean style nonetheless (Jones, Levi, and Bettelli 2005, 542; Vagnetti 1999, 144). The popularity of chinoiserie in eighteenth- and nineteenth-century Europe illustrates the same point (Impey 1977; Jacobsen 1993). The fashion for Oriental objects and styles at this time led to the production of the whole new class of items known as chinoiserie. These items were made to suit European tastes and bore little resemblance to genuine Oriental styles, but they nonetheless contained enough visual cues to evoke Oriental associations for most European consumers. The need to recognize innovation in stylistic imitation therefore argues against adhering too strictly to traditional stylistic typologies. It is certain that for studies of intercommunity interaction, the extent of deviation from formal stylistic criteria is crucially important, as it can be indicative of the frequency and extent of contacts. However, for studies of community identity, it is more important that a community considered a style as having external associations. Recent archaeological studies have highlighted the importance of social meanings at the point of consumption, pointing out that the meaning an object may have held for its consumer may have been radically different from those attached to it at Page 52 →the time of its production or at other stages in its life history (Appadurai 1988; Hodos 2006, 8). Especially in situations of intercommunity contact, objects can accrue new meanings, forged in the active process of cultural encounter and hybridization (Dietler 1999, 483; Hodos 2006, 9). For an object to signify an external connection, therefore, it need not necessarily have been either an actual import or an imitation bearing strict stylistic similarities to the original external style. It could simply have recalled the general aesthetic associated with that external style and given off visual cues that prompted the viewer or user to slot it in with that style. An object can denote and connote an external connection not just by belonging to an external style but also by merely evoking that style. As argued in both archaeological and psychological theory, stylistic recognition works on the basis of unconscious comparisons to slot an artifact into a mental network of social meaning (Gamble 2001, 198; Hegmon 1992, 519–22; Wiessner 1984, 191–95; Wiessner 1988). Such theories of style are related to the ideas of semiotics, which assert that signs and symbols only gain meaning as part of a wider framework of interrelating signs, a conceptual stance that has been applied to material culture by theorists within the fields of both archaeology and art history (Bal and Bryson 1991; Knappett 2005, 85ft; Tilley 1988, 185–88). When considering community identity, then, material culture styles must be considered in the context of the wider cultural experience within which they would have been viewed. Such an experience-based approach to material culture style, focusing on the social perception of objects within their wider cultural contexts, is in line with the “phenomenological turn” in archaeological theory (Hodder and Hutson 2003). The adoption of such an approach alongside more traditional typological ones would reconcile phenomenology, cultural experience, and psychological theories of stylistic recognition with traditional archaeological analysis. Whether an object was considered to be external or indicative of an external connection is therefore a complex question and cannot simply be determined from the object's place of production or adherence to a strict set of formal elements that define an archaeologically documented style. The cultural experience of those who encounter the object and whether this cultural experience leads them to classify it conceptually as external are also of central importance. Considering material culture styles in the context of wider contemporary cultural experience while still making full reference to traditional systems of artifact classification acknowledges that elasticity and cultural specificity must be built into such systems of perception and social meaning. It shifts our focus firmly onto the point of Page 53 →consumption, rather than production, and exclusively considers the social meanings associated with an object at its point of final use. Who Is the “Them”? Performative Representations Once objects and styles that are likely to have carried connotations of certain external groups have been identified, the next stage of the analysis must consider how these objects and styles were used and in which social situations.

We have already established that performative representations can portray external groups in a positive or negative light and as either closely related or remote and distanced. There is no clear rule for determining when and how symbolic objects are used to create an image of Otherness, however. Instead, each item must be considered as a separate case for the image it projects. Three main elements determine the nature of this portrayal: the physical characteristics of the object itself, the social situation in which the object is used, and the manner in which it is used. A formal silk kimono, for example, is likely to evoke associations with Japan for a European viewer and can therefore be used to construct images and perceptions of Japan through performative representations. The nature of these images will partly depend on the physical characteristics of the kimono—a high-quality silk kimono will suggest different associations from a cheaper cotton model, for example. The social situation in which the kimono is used also informs the nature of the overall image constructed: if it is worn in a formal public setting such as a graduation ceremony or a state banquet, the connotations may be quite different from if it is worn in the bedroom. Finally, the way that the object is used within this social setting is also important. Wearing the kimono at a state banquet suggests something different from displaying it as a wall hanging at a state banquet, and using it as a napkin or dishcloth on the same occasion would imply something very different once again. Therefore, the physical characteristics of an object, the social setting in which it is used, and the way in which it is used all determine the role of that object in the performative representation of the external Other. All of these three critical elements can be investigated archaeologically. First, the physical characteristics of an object have an impact on the kind of image of the external that object can be used to create. While an object may recall associations with an external group, it may also suggest connotations of rank and status, if it is made from materials considered to be particularly valuable or cheap. It may also evoke gendered ideas, if the object is closely connected to gender roles or is usually used by only one Page 54 →gender. Objects can therefore connote several different social ideas as well as that of the external connection, and these additional connotations will all reflect back on the image of the external group being created. The case of Mycenaean pottery found at Late Bronze Age sites in the central Mediterranean illustrates this. These vessels are likely to have evoked associations with the Aegean within a central Mediterranean context but seem to have had associations of high status on Lipari, as most of the Mycenaean vessels were fineware dining shapes (van Wijngaarden 2002, 251–52), while the import of Mycenaean coarse storage wares to Vivara may suggest fewer elite associations (Rizio 2005). Second, the social situations in which an object is used are also an important contributing factor to the overall image created. This point can be illustrated by returning once more to the example of Mycenaean vessels in the central Mediterranean. At Thapsos on Sicily, Mycenaean stirrup jars and alabastra were commonly used as mortuary offerings (van Wijngaarden 2002, 233–36). This would probably have fostered a slightly different image from that present at Broglio di Trebisacce, where the same vessels were found widely in domestic contexts. Especially relevant for the purposes of this book is the use of such objects in one particular social setting—in enactments of community. The choice of what objects to use in these situations and the external associations that these objects might recall are crucial factors in the construction of community identity. While the general use of objects with external associations in the wider community is therefore informative, the specific use of such objects during enactments of community is particularly important for an archaeological study of community identity. Third and finally, the manner of use of the object in the given social situation also contributes to the image being created. It should be noted here that the physical form of an artifact does not determine that object's use, much less its social meaning. The glass jam jar, for example, is designed primarily as a vessel for the storage of small quantities of foodstuffs. However, in modern European homes, jam jars are frequently used to hold pens and pencils, as containers for loose change, or as temporary pots for small houseplants. Physical form and even the manufacturer's initial intent does not necessarily dictate the social use of an object. Only the archaeological context in which an object is found can establish how the object was used at the point of deposition. Nonetheless, an object's physical characteristics do impose some fundamental parameters on the range of ways in which it could potentially have been used. A jam jar does not, for example, make for a very effective tool for cutting bread, nor is it Page 55 →a suitable container for bulky foodstuffs such as watermelons or cuts of meat. Therefore, while the use of an object is not predetermined by its physical nature, it is somewhat related to it—social use is partly

influenced by the basic “affordances” and “constraints” of physical form (Gibson 1977; Knappett 2005). If an object is symbolic of an external group, its physical characteristics, context of use, and actual usage all contribute to a symbolic and performative representation of that group. This representation can cast the group as being socially accepted and familiar; or it can cast the group as being Other, socially distanced, and conspicuously “them” as opposed to “us.” No universal set of criteria can be advanced for determining when a performative representation casts a certain group as “them,” as a complex set of different factors is always at play in every instance. Instead, individual objects and styles must be considered separately to discern the image that they project of the people and places they represent. The Social Rationale of “Them” Once it has been established which external groups are cast as them through the symbolic use of material culture, some conclusions can be drawn about the nature of the community's social rationale and the social criteria that define community membership. This rationale determines which other geographic communities will be considered as truly Other and which will be considered as more familiar. The key to understanding the social rationale of community identity lies in the wider social and historical context. This includes both the internal dynamics of the community itself and also the wider contemporary circumstances in which it is situated. As community identity is both inward- and outward-looking, it gains social meaning from what is happening both within the community and without. The social relationships among community members and the interactions between groups and individuals within the community are perhaps the most obvious places to start when gauging the social context of community identity. The social structure of the community, the to-ings and fro-ings of status negotiation, the extent of internal stratification, and the intensity of power politicking are all relevant issues. Also relevant are the economic imperatives and limitations acting on a community—whether economic surplus is available for social investment and the proportion of the community that had access to this surplus. In addition, the level of craft specialization and the main areas of economic activity in the communityPage 56 → are of relevance. Social mores also contribute to shaping what circumstances are deemed socially significant and therefore suitable for symbolic signaling and which objects were thought to have been most suitable and significant for signaling with. It is imperative, therefore, that what is happening within a community and the social dynamics of the individuals and groups that make up that community are taken into account when interpreting the way in which the identity of the community as a whole is expressed and articulated in material culture. The wider intercommunity context beyond the group in question is also of crucial significance. The more general historical developments at a regional or interregional scale play an important role in determining how any given community perceives itself and its identity in relation to others. Levels of external contact, whether driven by internal or external forces, have an impact on community identity. So, too, do patterns in trading and the fluctuations of consumer demand that can influence long-distance trading. The changing landscape of interregional politics, major historical events, rivalries between states or principalities, political domination or dominance, and military and diplomatic engagement are all potentially significant factors. Against this background, the evidence for which groups were treated as them and which groups were accepted should tell us something about the social rationale of the community. Using contextual information, it should be possible to interpret why some people were symbolically embraced while others were held at arm's length. Community identity must therefore be seen in its wider historical context, both within and without the community, for it to be possible to infer its social meaning. To return to a relatively modern-day example, it would not have been possible to conclude that the identity of English coal-mining communities was linked to professional occupation without contextual knowledge of the economic activity both within the community (the employment of a substantial proportion of the working population in the mining industry) and without (that not all contemporary English communities had the same patterns of employment and that there was a widespread economic demand for coal in general). For such wider historical contextualization, literary and oral evidence are obviously very helpful where available (Bell and Newby 1971, 166–71). In the absence of such testimony, such contextualization can be more difficult. In the case of Late Bronze Age and Iron Age western Anatolia, this contextualization must come

from a variety of sources, including archaeological remains from locations other than the case study sites, Hittite and Near Eastern documentary records, and later Page 57 →literary accounts. These sources are limited in scope by their nature and cannot offer a fully contextualized picture of the environment within and without the case study sites. They do, however, offer a place to start (see chapter 6 for further discussion of this).

Community Identity and Material Culture The previous three chapters established a theoretical model for community identity in archaeology. Archaeology should understand the community as a socially constructed form of group identity, negotiated through social practice and specifically through enactments of community. These enactments of community encourage a sense of collectivity by playing down differences between members and emphasizing commonalities instead. One of these key commonalities is rooted in a firm sense of place and the shared experience of coresidence. However, as community identity necessarily contains both spatial and social aspects, this emplaced sense of identity is often also overlaid with other forms of social identity, such as ethnicity, professional occupation, and religion. These additional social factors give the geographic community its social rationale, and it is against this social rationale that the Other of community identity is defined. According to this social rationale, groups that lie beyond the social as well as the geographic boundaries of the community will be symbolically treated as Other. This chapter has attempted to lay out a practical methodology for investigating community identity through material culture. Textual, ethnographic, and artistic sources are all useful tools for investigating community identity but may not always be available in all circumstances. An archaeological approach to community identity must therefore depend primarily on material culture. It is proposed, first, that material culture can bear the traces of enactments of community. This usually takes the form of evidence for large-scale social events, staged in a suitably open and undifferentiated place and making use of material culture that is relatively standardized. This allows for the physical gathering of many people and encourages group cohesion and solidarity by glossing over social differences and individual identities. Archaeological material can therefore provide evidence for enactments of community and a sense of collective us. Material culture can also offer clues as to a sense of them and the social rationale of community identity, especially in the way it is used to create Page 58 →performative representations of the external Other. First, objects that may have recalled associations of the external Other can be identified in the archaeological record. It is not necessary for an object to be a genuine import or to be made in a genuinely imitated foreign style—an object merely has to give off enough visual cues to evoke a sense of the external Other. Once these objects have been identified, it is then possible to consider how they may have been used to create a particular value-laden image of external Other. The way such objects were used is partially constrained by their physical form but can only be fully inferred from the archaeological contexts in which they are found. Relevant here is not just the manner of use but also the occasion for this use. The social situations in which an object is used also imbue it with meaning. For the purposes of this book, the use of such objects in enactments of community is particularly important, as this is when a sense of collective community us is negotiated in opposition to the materially represented external them. The choice of who to cast as them can be interpreted in the wider social and historical context to draw conclusions about the social rationale of the community identity. The theory and method presented in the first five chapters of this book offer a tentative framework for thinking about community identity in archaeology. The remainder of the book tests and elaborates on this framework, considering community identity at two case study sites in Late Bronze Age and Iron Age western Anatolia. Both of the archaeological sites selected show evidence for community identity at different times in their histories and for different reasons. In each instance, the relative salience of community identity at different times will be investigated. The community identities at Aphrodisias and Beycesultan will then be examined in their wider social and historical context, looking at each community's internal social dynamics and their wider regional and interregional situation. This process, it will be shown, points to some unexpected conclusions, not just about the communities in question themselves, but also about western Anatolia as a whole and the wider Mediterranean.

Page 59 → CHAPTER SIX

Overlooked Communities Late Bronze Age and Iron Age Western Anatolia

A Place In-between Chapters 1-5 of this book have established the concept of community identity as a focus for archaeological study and considered the ways it might be investigated through the archaeological record. Chapters 6-8 will put these ideas into practice, examining the changing nature of community identity at two specific locations in western Anatolia—Aphrodisias and Beycesultan. Western Anatolia has already been introduced in the first chapter of this book as a place considered to be in-between—between Europe and Asia, the East and the West, the classical world and the Near East. All too often, the region is discussed in terms of the grand narrative of culture clash; and far too frequently, the conventional literature characterizes western Anatolia as marginal, the periphery to someone else's core. This makes western Anatolia a particularly appropriate location within which to investigate community identity. Adopting a community identity approach in a region usually considered to be in-between offers an opportunity to consider the complex local dynamics of the region for the first time, focusing in on the people who lived and died there, rather than those who merely passed through. The geographical region known as western Anatolia is defined as lying between the Aegean coast and the central Anatolian plateau, with the dividing line between it and central Anatolia roughly adhering to the line that can be drawn between the modern cities of Antalya and Bursa. The landscape of western Anatolia is characterized by ecological diversity and includes coastal plains, river valleys, and mountain ranges. In this, it differs markedly from the landscape of central Anatolia, which mostly comprises a flat plateau of more or less uniform high altitude, flat plains, and Page 60 →upland ecology (Dewdney 1971, 151ft). The physical topography of western Anatolia is generally arranged around the long river valleys running on an east-west axis, connecting the coast and the central Anatolian plateau. Despite this apparent east-west trend in the geography however, there is also considerable scope for more complex patterns of interaction around the region—along tributaries, around plains, and over mountain passes (French 1981; Robert 1962; Thonemann 2006). But the preponderance of river valleys and the accompanying east-west routes has meant that a lot of emphasis is placed on the region as a corridor of communication between the Aegean and central Anatolia—a transitional area caught between cultures—rather than as a dynamic and independent region in its own right. This is evident from the way the region is currently viewed in the popular and the political imagination. Western Anatolia, the tourist hub of modern-day Turkey, is particularly characterized as a marginal and peripheral region, caught between the East and the West, Western capitalism and Islamic fundamentalism. This contemporary conception of the area as liminal and between has been highlighted in the political sphere recently by the controversy over the opening of talks discussing Turkey's entry into the European Union, where heated public debates over the status of Turkey were carried out openly and often vociferously in the European media (e.g., Hindle 2005; Lake 2005; Rumelli 2008). This betweenness is also felt within Turkey, especially in the discourses surrounding Turkish national identity (Kaya 2004, 57-67; Mango 2004, 233-42). Betweenness is also central to the way the region is represented in the sphere of culture and entertainment and to the way it is understood in the popular, as well as the political, imagination. For example, the Lonely Planet guide to Turkey claims that it “remains at the heart of the ideological battle between East and West” (Yale et al. 2005, 25). In a similar vein, reviews of the Nobel Prize-winning author Orhan Pamuk's celebrated novel Snow identify its key theme as “Turkey's troubled love-hate relationship with the West” (McGaha 2004) and “Turkey's fragile, undetermined identity” (Evans 2004). But it is not just in the spheres of popular representation and current politics that western Anatolia is seen as a transitional border zone between two other cultural regions. Historical and archaeological studies of its ancient past have also characterized the region as liminal, describing it as falling between the maritime societies of the

Aegean and those of the Anatolian plateau and being caught between these two larger political and cultural spheres of influence (e.g., Briant 1996, 682-83; Bryce 2006, Page 61 →77-86; Hornblower 1983; Lloyd 1956, 6ff.; Ma 1999, 43-101; Marchese 1986, 24; Mcqueen 1996, 39; Mee 1998; Neimeier 1999; Seeher 2005; Shipley 2000). For example, in a general essay on “Ionia,” the ancient historian Hogarth explicitly essentializes the region as caught, throughout its history, between “a Power of Europe” and the “continental Powers of Asia” (1909, 47-48). The pioneering archaeologist Lloyd similarly saw Anatolia as a “land-bridge” between Europe and Asia (1956, 7), with western Anatolia in particular necessarily having to be drawn into either the world of the Aegean or that of the central Anatolian plateau (1956, 6). Even recent studies of the region tend to essentialize it as fundamentally between, with descriptions such as “a contested periphery” (Cline 2008) or “a peripheral area between specific geographic and geological blocks and cultural zones” (Marchese 1986, 24). Though some archaeologists have recently sought to problematize this characterization and highlight the intrinsic complexity of the region (Greaves 2007; Özdo an 2007), the characterization remains. The idea of Anatolia as being caught “between two distinct cultural and geographical spheres” (Greaves 2010, 27) is proving remarkably hard to shake. For some periods of antiquity, this characterization of betweenness is justified. In the classical period, for example, western Anatolia was the location of intense political conflict between the Achaemenid Persian Empire and various Greek groups (Briant 1996, 682-83; Hornblower 1983); and in Roman times, it was a theater for competition between Roman and Seleucid imperial interests (Ma 1999, 43-101; Shipley 2000). Nonetheless, the consistent tendency to consider the region in terms of betweenness in both its present and its ancient past cannot be explained simply with reference to specific historical situations such as these. Nor can it be explained by the physical geography of Anatolia, as the east-west characterization of the region is only one of many socially constructed perspectives on its geography. Such characterizations depend on the adoption of a point of view external to Anatolia—an outsider's angle that defines the region as peripheral rather than central. The prevalence of this perspective ultimately derives from a set of historically specific mental categories—the East and the West, Europe and Asia—which define western Anatolia as a border region lying somewhat uncomfortably between them. The conceptual development of this system of geographical-cultural classification was famously highlighted by Edward Said in his 1978 book, Orientalism, which traced the intellectual construction within western European society of the East, or the Orient, as the exotic, mysterious, and barbarous counterpart of the West. The creation Page 62 →and application of these categories, Said argued, both conditions social perceptions and informs prevailing social attitudes. However, lying along the fault line between the perceived geocultural blocks of Asia/Europe and East/West, the western part of the Anatolian peninsula is difficult to fit into this binary scheme. The region has therefore been seen as between two worlds, incorporating elements of both “us” and “them,” self and other. This has led to the region's widespread characterization in terms of betweenness, a cultural crossroads, and culture clash. This perception of the East and West as fundamentally opposed and the consequent characterization of western Anatolia as falling into an uncomfortable taxonomic no-man's-land somewhere between the two are evident in the boundaries between academic disciplines still in use today. While the old and venerable discipline of classics was focused primarily on Greece (Dyson 2006, 1ff.), the almost equally old and venerable tradition of Near Eastern study was mostly concerned with the Bible lands, including Egypt, Mesopotamia, Syria, and the region of modern Israel and Palestine (Matthews 2003, 1-18). Anatolia has always lain somewhat between these two disciplinary poles, as biblical references to the Hittites pulled it toward the Near East at the same time as the idea of Hellenic Asia Minor pulled it toward the classical world. The turn of the twentieth century saw the full integration of central and eastern Anatolia into the sphere of the Near East, with the opening of excavations at the Hittite capital city Hattu a at Bo azköy and the discovery of Neo-Assyrian remains in southeast Anatolia (Gurney 1964, 1-14). Western Anatolia, by contrast, remained a region resolutely between. Over the course of the twentieth century, both fields diversified within themselves but remained distinct from each other, with classics, classical archaeology, and Aegean prehistory in one disciplinary camp and with Hittite archaeology, cuneiform and hieroglyphic scripts, and connections with the Bible lands in another. Western Anatolia, however, was never fully incorporated into the main research areas of either of these historiographical disciplines, remaining instead very much on the margins, peripheral to both and fully part of neither.

Marginality, edginess, and betweenness are therefore key characteristics of western Anatolia's historiography, perhaps more than of its history. Western Anatolia is a region that is crucially important for our conception of the wider Mediterranean world but poorly understood on its own terms. The following chapters of this book are an attempt to go some way toward redressing this balance, focusing on the people of western Anatolia themselves. The community identity approach offers an opportunity Page 63 →to examine different groups within western Anatolia not in terms of what external influences were imposed on them but in their own right. The analysis will consider the dynamics within two settlements in the region, relating these back to the wider historical context around them. At what stages and for what reasons did people at these settlements ascribe to a conscious sense of community identity? Why might they cease to do so at some later stage? When and why do the social dynamics differ between the two settlements? The answers to these questions will offer new insights both into western Anatolia as a region and also into the Late Bronze Age and Iron Age (hereafter LBA and IA) period. First, however, a historical overview is necessary to set the scene for the detailed case studies that are to follow.

Historical Overview Scope of the Study and Sources Writing was known for much of the LBA and IA, and some limited textual evidence is available. However, epigraphic evidence is scanty and can be found at only a small number of sites in the region, and documentary records for much of the period have only been discovered in neighboring regions around western Anatolia and mostly focus on a relatively narrow range of subjects. For this period, therefore, western Anatolia is primarily, although not exclusively, accessible through the archaeological record. Textual evidence becomes much more widely available after ca. 800 BC, with many more inscriptions surviving and with literary sources from the region being preserved in the ancient Greek textual tradition. This remains the case throughout the long IA right up until the start of the classical period ca. 550 BC. This is roughly the date of the 547 BC Persian conquest of western Anatolia (Briant 1996, 45-47; Herodotus 1.86), a historical event that marks a turning point in the region's history. Until this point, the region was comprised of many politically independent units, interacting both with each other and with other regions. It is therefore particularly suitable for an experimental investigation of community identity and for testing the methodology set out in the previous chapter, as the independent local groups that covered the western Anatolian landscape were free to enact their own independent community identities. After the invasion of the Achaemenid king Cyrus I, however, these disparate units were subsumed under a larger imperial structure almost Page 64 →continually through the Achaemenid to the Seleucid, Roman, Byzantine, and finally the Ottoman empires. The ca. 550 BC watershed is therefore a socially meaningful point in western Anatolian history. It is also, however, a disciplinary horizon. The pre-Achaemenid period of western Anatolia falls into a disciplinary gap between prehistoric archaeology and classical archaeology/history. Similar disciplinary gaps are evident for the same period elsewhere in the Mediterranean (Hodos 2006, 3-4), but the gap is felt particularly strongly in the study of western Anatolia. The opening of the classical period and the start of Achaemenid suzerainty over the region is therefore taken as an appropriate point at which to halt these case studies. As mentioned earlier, written evidence from LBA-IA western Anatolia is extant but scanty. No archives of documentary evidence dating to this period have yet been found in western Anatolia itself, although extensive archives elsewhere in the eastern Mediterranean can sometimes shed light on the region. Hittite documents from the imperial capital of Hattu a in central Anatolia during the LBA (ca. 1500-1200 BC) contain valuable information about how the Hittite administration characterized the region and its dealings with it. Contemporary documents from the Mycenaean Aegean written in Linear B may also contain references to western Anatolia, but toponyms and ethnonyms in Linear B are not yet fully understood, and it is unclear how much can reliably be inferred from these texts, which are brief and deal primarily with palace administration. More LBA documentary references to western Anatolia can be found in two letters from the diplomatic archive of the pharaoh Akhenaten at Amarna in Egypt. Moving forward in time to the Early Iron Age (ca. 1200-800 BC; hereafter EIA), there are some references to the western part of Anatolia in the Neo-Assyrian documents of the ninth and tenth centuries BC, although these are rare because the region lay outside the Neo-Assyrians' immediate sphere of contact. Slightly later still are the literary texts in ancient Greek written by the Hellenic inhabitants of western Anatolia's

Aegean coastline and preserved in the classical textual tradition, including works of poetry, philosophy, geography, political and natural science, and historiography. While information from all these sources is invaluable for contextualizing western Anatolia in the LBA-IA, it should be borne in mind that these texts record not the perspectives of western Anatolians themselves but, rather, the external perspectives of outsiders on the region. The Greek texts are a notable exception to this, although their preservation through Page 65 →the manuscript tradition means that they do not offer an unbiased sample of the range of texts produced at the time, as they were selected for preservation by librarians and copyists over the centuries. While they do express some of the options held by western Anatolian populations, their status as documents written by the literate elite means that they cannot be taken as representative of the full range of points of view that were current at the time (Finley 1985, 14-16; R. Thomas 1994). All of these texts—Greek, Hittite, and Neo-Assyrian—will be drawn on in the following analysis, and all of their specific perspectives will be taken into consideration. The corpus of epigraphic material from this period in western Anatolia is also relatively limited, but fortunately is steadily growing. Monumental Greek inscriptions in stone are well known for the final phases of the period (see the Tituli Asia Minoris volumes as well as the various publication of inscriptions from individual sites). Monumental stone inscriptions are also known in hieroglyphic Luwian, Carian, and Lydian, although these are currently few in number and are often not lengthy texts. Notable among these are the Karabel inscriptions in hieroglyphic Luwian (Hawkins 1998). Inscriptions are also found on a smaller, nonmonumental scale, and there are larger numbers of potsherds, seal stones, and seal impressions known from a variety of sites inscribed in Greek, Carian, Lydian, and Phrygian (Brixhe 2004; Brixhe and Lejeune 1984; Hawkins and Easton 1996; Günel and Herbordt 2010; Lajara 2007; Melchert 2004a, 2004b). Our understanding of western Anatolian scripts and languages is constantly improving, especially for the latter part of the case study period. However, there are still very few inscribed sources dating from the early part of our period. Archaeological evidence, on which any investigation of LBA-IA western Anatolia is primarily dependent, has increased in both quality and quantity over the last few decades, as more and more excavations have chosen to take early levels into account alongside the more eye-catching remains of the classical and Roman period and the wealthier deposits of the Neolithic period and Early Bronze Age. The last thirty years have seen many new projects specifically aimed at examining the prehistoric or the protohistoric periods, as well as a number of regional surveys. These new projects have significantly added to our understanding of western Anatolia. However, despite the substantial increase in the archaeological material available for studying western Anatolia during this period, the sum total of information available in recent decades remains small in comparison to that available for either of its neighbors that have Page 66 →received more intensive study, central Anatolia and the Aegean. Large areas of western Anatolia remain unexplored archaeologically, and these substantial geographical blind spots impede any attempts at comprehensive regionwide overviews. The Late Bronze Age During the LBA, western Anatolia seems to have been a fluid patchwork of independent polities, territories, and communities, linked both to each other and to groups beyond the region by networks of trade, communication, and diplomacy. Contemporary documents of the Hittites tell us that the region was generally known to them as “Arzawa” (fig. 2). This designation seems to have been quite flexible in its use, at some points referring to a specific political entity, at others being used as more of a generic geographic term. In any case, the documents suggest that there were a number of political units in the region, which were regularly combining, fissioning, and reconfiguring themselves (Garstang and Gurney 1959, 75-109; Hawkins 1998, 2009; Heinhold-Kramer 1977). We also read of alliances being made and broken, raids being carried out, and secret deals being struck. Perhaps the example par excellence of this behavior is that of the ambitious Arzawan ruler Madduwatta. Madduwatta took over a number of Arzawan states and annexed several Hittite territories, all the time building up a secret web of potentially conflicting alliances, including pacts with Pitassa and Dalawa as well as the Hittites. His exploits are recorded in a text known as “The Indictment of Madduwatta” by Arnuwanda I, a lengthy complaint on the part of the Hittite king, listing examples of Madduwatta's acts of double-dealing and treachery (Beckman 1996, no. 27).

The Hittite Empire was also involved in the complex politics of the region, mounting campaigns against polities by which it felt threatened, signing treaties with local rulers, and using its influence to try to stabilize the political and dynastic flux. Crucially, when Uhhaziti of Arzawa managed to form a pan-Arzawan confederacy in the early thirteenth century, Mursili II led a successful campaign into western Anatolia and divided Arzawa into three main vassal states: Hapalla, Mira and Kuwaliya, and Appawiya and the Seha River Land (Beckman 1996, nos. 10-12; Grélois 1988, 58-66, 677-83). The texts also mention a state to the west of the region, Ahhiyawa, whose location has been variously identified as somewhere in Mycenaean Greece, the Dodecanese, and Thrace. Regular interaction between the elites of Arzawa and Ahhiyawa is implied in much the Page 67 →same fashion as interaction between various Anatolian elites, involving alliances, wars, and dynastic marriages, all conducted with a healthy measure of duplicity and backhandedness (Singer 1983). It is indicative of the academic marginalization of western Anatolia that Ahhiyawa and the as-yet-unresolved “Ahhiyawa Question” have attracted far more scholarly attention than the Arzawan states. Much of this attention is because of the likely connection of Ahhiyawa to the Mycenaean sphere, as well as the associations frequently made with the Trojan War (e.g., Benzi 2002; Finkelberg 1988; C. Gates 1995; Güterbock 1983a; Hope Simpson 2003; Huxley 1960; Mcqueen 1968; Mellaart 1982; Mountjoy 1998; Neimeier 1999; Pavúk 1995; Sommer 1932).

Generally, the records claim a high level of Hittite control over the region, using the standard language of vassalhood, subordination, and obligation with reference to the Arzawan states. In addition, there is some possibility that the Hittite administration had economic interests in this area, especially in the production of purple dye (Singer 2008). However, the documents are not impartial historiography but carefully constructed imperial rhetoric and should not be taken completely at face value, as they are instruments of imperial self-fashioning as well as documentary Page 68 →records (Güterbock 1983b). Overall, it does not seem that western Anatolia was ever properly part of the Hittite Empire in the same sense as other areas such as northern Syria and the Amuq (Glatz 2006, 349; Mellaart and Murray 1995, 108; Sagona and Zimansky 2009, 285; Seeher 2005, 40). For a start, the unusually frequent references to local rulers ignoring or openly flouting Hittite demands suggests that the situation on the ground was very different from the picture presented in the official rhetoric. In addition, the Karabel inscriptions and monumental relief sculptures carved into a cliff face of a pass across the Tmolus range on the route between Ephesos and Sardis are evidence that there was at least one long-established and stable royal dynasty in the region, one that David Hawkins has recently argued may have presented quite a threat to the Hittite rulers in the final years of the empire (Hawkins 1998, 2002). Finally, direct contacts between Arzawan rulers and the kings of Egypt suggest more independence on the part of western Anatolian leaders than the Hittite texts concede. For example, Amenophis II of Egypt entered into negotiations with Tarhundaradu of Arzawa to marry an Arzawan princess, and Arzawan kings were greeted by the pharaohs with the standard diplomatic formulae usually reserved for communications between political equals (Hawkins 2009; Moran 1992, nos. 31, 32). Overall, the emerging picture of western Anatolia is of a region within the Hittite sphere of interest and influence but not within the empire, independently interacting with groups across the eastern Mediterranean as well as with its neighbors immediately to the east and west, and comprised of several politically autonomous units that had some sense of coherent social or political identity at certain points during the LBA. Evidence from the archaeological record largely supports this picture, with the material culture of the period showing some similarities, but not uniformity, across much of the region, incorporating the material traces of contacts with many parts of the eastern Mediterranean. Each site so far uncovered appears to have its own independent ceramic development, although certain features and shapes recur in several assemblages across the whole region, such as the omnipresent beaked jug, best known from Troy, and the use of metallic slips for finewares, most commonly attested at Beycesultan. Somewhere between these very general regional patterns and the overall impression of site specificity, there are also some indications of cultural zoning within the region. The Mycenaean-influenced area on the southern Aegean coast and nearby islands is the most fully discussed of these cultural zones, yielding substantial amounts of its own regional style of Mycenaean finewares and Page 69 →incorporating several elements of Mycenaean material culture and social practices better known from the Greek mainland and Crete. Conspicuous mortuary consumption in tholos tombs, for example, is known from Panaztepe, Ephesos, Colophon, Müsegebi, and De irmentepe near Miletus as well as on the nearby Dodecanese islands,

including Rhodes and Cos, and is often associated with items such as Mycenaean-style swords as well as pottery (Georgiadis 2003; Mee 1998, 138-41). The prevalence of Mycenaean-style material culture in this zone, which Mountjoy calls the “east Aegean-west Anatolian interface” (1998), precludes considering them as an external influence; they are simply part of local elite material culture. Needless to say, this engagement with the material language and ideology of Mycenaeanness by communities in this coastal and island zone suggests close contacts with the rest of the Aegean. Given our imperfect understanding of what being a Mycenaean may have actually meant (Bennet and Davies 1999), we should consider this zone as having its own geographically situated cultural forms engaging with wider Aegean developments, rather than derivative forms based on Mycenaean “influences” (C. Gates 1995). Another zone with some broad similarities in material culture and social practice is the northwestern area, including the Troad and the islands of Lemnos, Lesbos, and Samothrace. In the LBA, this area shared many similarities in coarse wares, as well as the eponymous Northwest Anatolian Grey Ware (Bayne 2000). Basedow has also pointed out that there is a broad similarity in the mortuary traditions of this northwestern zone, which she suggests may indicate some conscious sense of unity through deliberately reproduced social practice (2002). The final zone that has been tentatively identified is that of the inland southwest, in the area of the case study sites considered in the following two chapters. This zone was focused around the Maeander river valley and included such sites as Aphrodisias, Beycesultan, Laodicea, and Kusura. The pottery of this area is characterized by a preponderance of metallic slips and a tendency toward metal skeuomorphism. This inland southwestern zone is described in more detail later in this chapter. There were therefore three potentially meaningful cultural zones in the LBA: the south Aegean coastal, the northwestern, and the inland southwestern. However, given the patchy nature of the archaeological record, the social meaning, nature, and extent of these cultural zones are still unclear. While there have been some attempts to interpret these areas of tenuous cultural similarity as states or ethnic groups (Basedow 2002, 474) and some effort to link both sites and locales to toponyms in the Hittite texts (see the studies on the “Ahhiyawa Question” previously cited), Page 70 →the current state of the evidence means that archaeology can neither confirm nor deny such speculations. The shallow cultural similarities mentioned here may not have translated into any socially meaningful sense of identity. In addition, as mentioned earlier, each western Anatolian site so far uncovered appears to develop along its own cultural trajectory, and no large-scale social identities spanning these broadly defined cultural areas can be inferred at this stage. As well as illuminating the dynamics within the region, the archaeological record also bears witness to contacts beyond the region. The standardized range of Hittite ceramics, a commonly used index of imperial control in eastern and southern Anatolia (M.-H. Gates 2001; Gunter 2006; Henrickson 1995, 82-83; Postgate 2007), is completely absent from western Anatolia, thereby supporting the image of the region as politically independent. While a number of shapes at Beycesultan and Troy have been identified as having Hittite comparisons, the overall level of Hittite influence on the region's material culture is generally small. Some influence can be seen, however, in certain aspects of elite culture, such as the Karabel inscriptions already mentioned, which were written in hieroglyphic Luwian and accompanied by a Hittite-style relief using standard royal iconography (Hawkins 1998). This central Anatolian influence on status items is also evident on a bronze stamp seal from Troy bearing two hieroglyphic inscriptions (Hawkins and Easton 1996); a seal impression with hieroglyphs from Çine-Tepecik (Günel and Herbordt 2010); and a stone stamp seal found at Beycesultan, bearing a ladder-and-griffin motif (fig. 9; Cat. no. B2-53 in appendix A). The pattern of limited central Anatolian influence at specific locations, restricted to the elite sphere and the iconography of power, suggests neither Hittite control over the region nor intense interaction with central Anatolia at a grassroots level. Instead, it implies that certain western Anatolian elites were engaging with, adopting, and appropriating the Hittite visual language of power in a limited way. Western Anatolian elites also drew on the status symbols from elsewhere in the eastern Mediterranean in the form of imported luxury objects. For example, scarabs and faience beads from Egypt have been found in the necropolis at Panaztepe, one bearing the name of Amenophis III and another with the cartouche of Tutmosis III (Erkanal and Erkanal 1986, 73). The active use of exotic status iconography and luxury imports by the elites of the region reflects a wide geographical range of contacts, in terms of objects, ideas, and practices.

The coexistence of bulk commodity trade alongside such elite contacts Page 71 →is known for much of the eastern Mediterranean and has generated a vast amount of scholarly literature. Treatments range from studies of the material and written evidence for this long-distance trade (Cline 1994) to considerations of the driving social and economic forces behind it (Sherratt 1999, 2003; Sherratt and Sherratt 1991, 1993; van Wijngaarden 1999), as well as many regional studies (e.g., Gitin 1995; Heltzer 1988; Vagnetti 1993) and a number of edited conference volumes on the subject (e.g., Cline and Harris-Cline 1998; Davies and Schofield 1995; Gale 1991; Laffineur and Greco 2005). Given its location and the evidence for the bulk production of textiles at the site (Guzowska and Becks 2005, 282-85), involvement in such trade has been hypothesized for Troy (Easton et al. 2002, 101-6; Korfmann 1996, 94-95). However, any such trade has left little trace on western Anatolia itself. While there is some direct evidence for the region's involvement in such trade, such as the examples of Northwest Anatolian Grey Ware pottery in Cyprus and the Levant (Heuck Allen 1991), this involvement mostly has to be inferred from what we know of the trade itself, perhaps best illustrated by the Ulu Burun shipwreck found off the southern coast of the region (Bass 1987, 1991). However, generalizations on a regional level are of limited use in the fragmented territory of western Anatolia, and the issue of trade and contacts will be addressed on the scale of the individual community for the case studies under consideration in this book. The Early Iron Age The end of the Bronze Age in the eastern Mediterranean is traditionally seen in terms of collapse and decline. Previously thriving states and empires crumbled within a fairly short space of time (ca. 1200-1150 BC), and both the archaeological and documentary records for the period immediately after the collapses tend to be comparatively poor. In the Aegean, the Mycenaean palaces were destroyed and burned. In Egypt, internal social and economic problems were exacerbated by external attacks, including raid by the infamous “Sea Peoples.” In the Levant, major cities were destroyed, including Beth Shan, Megiddo, and the great trading emporium of Ugarit. Finally, in Anatolia, the Hittite Empire imploded under pressure from a series of rebellions, dynastic struggles, and external attacks. Conventional treatments dramatically declare the end of the LBA to be the collapse of civilization, labeling the following period as the “Dark Ages,” characterized by poverty, depopulation, and backwardness, lasting Page 72 →until the rise of the new literate states of the later IA (Barnett 1975b, 359; Coldstream [1977] 2003, 17; Cook 1975, 773; Desborough 1972, 12). This use of the term Dark Ages is a historical one, referring to the social darkness of the period under consideration. The term is more usually used now as a historiographical expression, referring to the relative lack of knowledge about the period, rather than as a social judgment. Indeed, the events at the end of the LBA have been reinterpreted in recent years (e.g., papers in Bachhuber and Robert 2009). Scholars such as Schoop suggest the gradual, rather than sudden, collapse of some state systems (2003), while others, such as Susan Sherratt, hypothesize endemic instability within LBA command economies and the growth of “fringe” economic activity (1998). It has also been generally acknowledged that many areas beyond the immediate core locales of the LBA states did not undergo a decline in the same way that previously important political centers did (Bell 2009). In the Aegean, EIA Lefkandi has emerged as a wealthy community with complex concepts of social differentiation (Popham and Sackett 1980). In Cyprus, at both Kition and Palaeopaphos, the end of the Bronze Age is accompanied not by decline but by an increase in urban redevelopment and monumental buildings (Karageorghis 1981, 105; J. S. Smith 2009; Steele 2004, 208-10). In western Anatolia, the end of the Bronze Age did bring some substantial changes, and in several instances, settlements were destroyed in what appear to be violent burning events. In cases such as Troy and Sardis, settlements were rebuilt to different plans around this time, while at the large majority of others, including Ephesos, Miletus, Bademgedi i Tepe, Panaztepe, Klazomenai, and Bayrakli, the poor state of preservation of the EIA levels has meant that continuing occupation of the site can only be inferred from finds of EIA pottery with no architectural context. Finds of Late Helladic IIIC (hereafter LH IIIC), Submycenaean, and Protogeometric pottery at coastal sites suggest some continuity in the south Aegean coastal zone, in terms of contacts with the western Aegean and of maintaining cultural similarity in the area (Lemos 2002, 211-12). However, in the northern and inland areas, there appears to be a greater divergence in the ceramic assemblages between sites than during the LBA, as well as noticeable changes in the assemblages of individual sites. At Troy, for example, Knobbed Ware

makes its appearance, bearing witness to increased connections between the community at Troy and continental Europe (Becks 2003, 49; Chabot Aslan 2009). Despite the increased cultural independence of sites, however, it seems that networks of trade and exchange did not completely break Page 73 →down. For example, Cycladic pottery and glass beads continued to be imported to Sardis at around this time (Spier 1983), and Troy's increased evidence for connections to Bulgaria have already been noted. The economic stability and social demand for exotic objects necessary for such long-distance trade are self-evident, and increasing numbers of studies are showing continuity also in long-distance trading. Cyprus in particular seems to have been at the center of networks of trade stretching across the central and eastern Mediterranean (which, of course, included Anatolia), and the flourishing eleventh-century Cypriot trade with the central Mediterranean has provoked much academic interest (Karageorghis 1994; van Wijnaarden 2002). The overall general picture of western Anatolia during the EIA is therefore one of a region showing not only fewer signs of internal cultural similarity between sites and some signs of change in both social practices and material culture but also a certain level of continuity in both the vigor of local economies and networks of long-distance communications. As well as both their Aegean and their more far-flung eastern Mediterranean and Bulgarian connections, the EIA communities of western Anatolia would also have been in contact with the expanding state of Phrygia, based at Gordion on the central Anatolian plateau. The Phrygian state is thought to have been established some time shortly after the fall of the Hittite Empire (ca. 1200 BC) and is said to have incorporated migrants from continental Europe as well as local groups. While little is known of the initial formation of Phrygia itself (the typical “darkness” of the Dark Ages obscuring such issues), it is evident that its influence steadily grew over the course of the eleventh, tenth, and ninth centuries BC. By the start of the eighth century, Phyrgia had incorporated the former Hittite heartland around Hattu a within its territory and also exerted a strong cultural influence over a wide geographical area, stretching from Daskyleion in the west to Carchemish in the east and Cappadocia in the south (Kuhrt 1995, 562ff.). Such influences can be seen in the wide distributions of Phrygian ceramics, cultic iconography linked to the Phrygian goddess Matar Kybele, and inscriptions in the partially deciphered Phrygian language found in a variety of contexts, from potsherd graffiti to monumental stone reliefs (Voigt and Henrickson 2000). However, in western Anatolia, the range of Phrygian influences was somewhat limited, and it seems that during the early first millennium BC, Phrygian interest tended to be directed eastward rather than westward (Liverani 1995; Sams 1993; Summers 1994). This is indicated by the strong Phrygian influences in east Anatolian /north Syrian material culture and the intense diplomatic activity Page 74 →between Phrygia, Urartu, and the Neo-Assyrian Empire, recorded in the Neo-Assyrian documents (Barnett 1975a, 418-26; Hawkins 1995; Kuhrt 1995, 562-66). On an interesting historical note, however, the same Phrygian king is mentioned in both the contemporary Neo-Assyrian texts and the later Greek tradition, appearing in the former as “King Mita of Mushki” (Ebeling et al. 1993, 271-73) and in the latter as Midas of the famed golden touch (Herodotus 1.14; see Vassileva 2008 for discussion). The strongest evidence for Phrygian influences in western Anatolia is at Daskyleion, a site that has no occupation history before the EIA but has Phrygian-style ceramics, tumuli, and cult remains, as well as graffiti in Phrygian (Bakr 1997). Apart from Daskyleion, however, any Phrygian influence on western Anatolia has left very few material traces, with some Phrygian pottery also found at Beycesultan (Mellaart and Murray 1995, 89-91) and Sardis (Ramage 1994) but none yet identified at any site on the Aegean coast. While they could be seen as indicative of direct political control over western Anatolia, these influences are relatively weak compared with those in southeastern Anatolia (Daskyleion excepted), so it might be more productive to think of local communities in western Anatolia adopting and adapting material culture associated with Phrygian elite status for their own internal uses. This certainly seems to have been the case on the island of Samos, just off the Aegean coast, where rhyta in the form of animal heads became popular votive offerings (Ebbinghaus 2008). As with the analogous example of Paphlagonia, there is a range of possible explanations for the presence of Phrygian pottery in western Anatolia, from Phrygian control over a network of key sites, for example, to trade between Phrygia and local political leaders (Matthews 2000, 35). Also at some stage during the early first millennium, Greek-speaking communities are thought to have settled on

the Aegean coast as part of the “Ionian colonization,” a phenomenon largely constructed on the basis of later Greek texts (including Herodotus, Thucydides, Pausanias, and Strabo) and considered by classical scholars to be the defining historical episode of the EIA. The colonization has been envisaged as a grand migration of people from the Greek mainland (with Athens leading the way) to the Aegean coast of Anatolia. The migrations were said to have caused the displacement of indigenous and emphatically non-Greek barbarian groups and to have involved the foundation of what later became the East Greek polis states of the later IA (Cook 1962, 23-35; Emlyn-Jones 1980, 10-17). The textual Greek tradition maintains that several distinct Greek groups migrated to Anatolia, led by mythic figures such as Neleus and Androclus, both sons of the Athenian king Codrus (Herodotus 9.97; Pausanias Page 75 →7.2.8-9). In some cases, the occupation was described as peaceful and resulting in a mixed Hellenic and non-Hellenic population (e.g., Erythrae; see Pausanias 7.3.7); in others, the texts describe it as violent and resulting in the wholesale massacre or expulsion of the site's previous occupants. At Miletus, for example, Neleus is said to have killed or driven out all the men and forcibly taken the women as wives (Herodotus 1.146.2-3), while Mimnermus speaks with regret of the settlers' bloody arrival at Colophon (fragment 12). Modern accounts of the Ionian colonization, however, tend to stress the divisions and conflicts between colonists and natives rather than mixing and cooperation. While the Greek texts themselves tend to tell foundation stories that imply a certain amount of mixing (according to Herodotus, even the Milesians are necessarily half-native through the maternal line), the historiography of the period written from the nineteenth century until the 1980s largely chooses to ignore Ionia's non-Hellenic elements. The coherence of the Ionian colonization has been somewhat overstated in modern treatments of the subject, and it is increasingly being recognized that the Greek texts offer a multiplicity of rival narratives concerning the Ionian colonization (Crielaard 2009). It is now widely accepted that such competing stories were a feature of the archaic Greek world and a means of claiming legitimacy, kinship, and authority and were therefore more the products of the complex sociopolitical environment in which they were written than the accurate reflections of a period long past (Calame 1990; Dougherty 1993; Greaves 2010; J. M. Hall 1997, 52; J. M. Hall 2002, 67-73; Hartog 2001; Malkin 1985; Osborne 1996, 32-37; Patterson 2010). Despite the fact that a single colonization horizon has now been questioned by sophisticated and historically situated readings of the texts themselves, archaeological evidence has often been sought to corroborate the conventional model of the Ionian colonization. The appearance of Submycenaean and Protogeometric pottery at coastal sites is often read as indicative of the arrival of the Ionians, with sites such as Miletus being identified as among the first wave of Greek settlements established around 1100 BC on the basis of a relatively small number of unstratified Submycenaean sherds. At the other end of the temporal scale, Bayrakli is thought to have been gradually settled by the Greeks over the course of the 900s, as Protogeometric wares appear in slowly increasing quantities alongside typically Anatolian wares (Akurgal 1962, 369-70; Cook 1958-59, 10-17). However, it has been noted that Aegean styles of material culture were already common in the coastal zone during the LBA and that Page 76 →this stylistic preference did not appear to undergo any radical realignment over the transition from the LBA to the EIA. In addition, Protogeometric pottery from coastal Anatolia can be shown to develop along its own regional trajectory, remaining clearly distinct from Attic Protogeometric styles (Lemos 2002, 211-12). The archaeological evidence therefore suggests that contact between the western and eastern Aegean should be thought of as being continuous, rather than starting in the EIA at a single horizon of Ionian colonization. Nonetheless, discussions around the Hellenic migrations to Anatolia still continue (see papers in Cobet et al. 2007). The emergence of a Hellenic identity in the coastal zone during the EIA seems likely to be due less to a mass migration than to a conceptual reorientation of identity within the coastal communities, who may have begun to think of themselves as Greek during this period. As in the comparable debate surrounding the Dorian Invasion on the Greek mainland, the Ionian colonization may have to do less with the “coming of the Greeks” and more with the “becoming of the Greeks” (J. M. Hall 2002, 45-47; Myres 1930). This is not to say that significant population movements did not occur at this time or that this period did not necessarily see a rise in the scale or frequency of migration. But contact, communication, and the movement of people around the Aegean were nothing new in the EIA and seem to have continued uninterruptedly from earlier periods. A reconfiguration of concepts of identity, self, and other therefore need not necessarily be dependent on the physical movement of population groups.

The Later Iron Age Such a reorientation of identity does seem to have happened, not in the EIA (as might be suggested by the conventional colonization narratives), but at some later “stage,” perhaps during the eighth century BC. The epic poetry of Homer, thought to have been composed at this time in either Bayrakli or on Chios, is indicative of this, focusing west to the Aegean rather than east to inland Anatolia. It has been argued that one of the main themes of the Odyssey is Hellenic identity (Osborne 1996, 155; Vidal-Naquet 1981), and the Iliad, although not specifically concerned with Hellenic identity versus non-Hellenic others, is still concerned, at some level, with the western Anatolian interactions with central and southern Greece (J. M. Hall 2002, 118; Konstan 2001, 31-32). Around the same time, typically East Greek styles of material culture—including pottery, architecture, and sculpture—were emerging, making the coastal zone culturally distinct from inland western Anatolia in a similar way as it was during Page 77 →the LBA. However, again as in the LBA, these local styles differentiated the area not only from the interior of Anatolia but also from the rest of the Aegean. East Greek styles of material culture were distinctly different from mainland Greek or central Aegean styles, suggesting the development of a specifically East Greek sense of identity that differentiated itself from the central and western Aegean just as much as from inland western Anatolia (Akurgal 1962, 377-78; Coldstream 1968, 262ff.; Cook and Dupont 1998; Crielaard 2009; Greaves 2010). By the seventh century, however, this sense of a Hellenic identity in the coastal zone had certainly crystallized, linked to the idea of Ionian origins and codified in the founding of the Ionian League (J. M. Hall 2002, 67-73). This appears to have been a loose federation of twelve independent city-states and served both to reaffirm the individual poleis in terms of territory and influence and to create a shared sense of identity based around putative origins and enacted in worship at a common sanctuary, the Panionion on Mount Mycale (Herda 2006; Lohmann 2006, 2007). Overall, the league's function seems to have been more ideological and social rather than overtly political, and it was not until 499 BC and the Ionian Revolt (after the temporal range of this book) that it assumed a significant role in interstate politics (Cook 1962, 34-35; Cook 1975, 803; Emlyn-Jones 1980, 17). It is significant that the Hellenic communities of both the north and south extremities of the coastal zone were excluded from the league, as this suggests that while the overall Hellenic identity discussed in the previous paragraph may have been important, Greekness was not the central concern in the founding of the league, which instead represented a much more selective group identity. Around the same time as self-consciously plural East Greek identities were emerging in the Aegean coastal zone, there is an increasing amount of evidence for contacts spanning western Anatolia from the coast to the plateau. Phrygian and East Greek styles seem to grow closer, with particularly similar trends in the decorative schemes used on pottery, woodwork, and textiles, from the geometric designs of the Early Phrygian period (Sams 1994, 175-96) and Greek Geometric (Cook and Dupont 1998, 15-25) to the animal style of the Late Phrygian period (Sams 1994, plate 97; Sams and Temizsoy 2000, 38-49) and the Greek Orientalizing style (Cook and Dupont 1998, 29-70). In addition to contemporary stylistic developments in material culture, there also seems to have been an increase in trade across western Anatolia. Central Anatolia and the Aegean appear in closer contact than ever before, with Phrygian objects turning up with increasing frequency even in the west and central Aegean after 800 BC Page 78 →and with East Greek items from the Aegean coast being imported in greater numbers to Gordion. Inland areas of western Anatolia also show the traces of this directional east-west trade, with items from both locales being found at Sardis, for example (Ramage 1994). Over the eighth century, inland western Anatolia did appear to have acted as something of a “between” or “middle” zone, incorporating elements from both the Aegean coast and the central Anatolian plateau into its material culture. This situation was to change over the turn of the seventh century, however, as the number of sites with burnt destruction layers from around this time suggests a degree of political instability in the region, perhaps linked to raiding by a group known from later Greek sources as the Cimmerians and from the Neo-Assyrian documents as the Gimirri. The Cimmerians are generally thought to have been a nomadic group or tribal collectivity and have yet to be firmly associated with any particular archaeological sites or material culture forms. This makes them hard to investigate, except from the external points of view of either the later Greek writers or the official NeoAssyrian governmental records (Ivantchik 1993; Kristensen 1988).

Probably at least partly due to the destabilizing activities of the Cimmerians, the influence of Phrygia declined toward the end of the eighth century and in the early seventh, a period that also saw the expansion of the Lydian state based at Sardis. The Lydians are well known from Greek authors such as Herodotus, who tell colorful stories of the expansion of their empire and the exploits, both political and amorous, of their kings (1.6-94). However, given their bias as Greek sources and the loss of works dealing specifically with Lydian history (e.g., the Lydiaca of Xanthos the Lydian, FGrHist 765), the trustworthiness of these sources is to be doubted (Kuhrt 1995, 567). The Mermnad dynasty is said to have come to power some time in the early seventh century, with Gyges deposing the last Heraclid king, Candaules, with the aid of Candaules's famously attractive wife (Herodotus 1.8-12). The gradual expansion of Lydian influence over western Anatolia during the course of the seventh century is attested by the Greek sources, which describe attacks launched by kings such as Alyattes against the Ionian cities on the coast, some of which have been linked to burnt destruction levels at sites such as that at Bayrakli (Akurgal 1962, 373; Cook 1958-59, 23-26; Strabo 14.1.37). The seventh century also sees the appearance of Lydian elements in the material culture of Aphrodisias (Mierse 1986), Bademgedi i Tepe (Meriç 2003, 86-87), Miletus (Greaves 2002, fig. 3.11), Bayrakli (Cook 1958-59, 17), and Daskyleion (Gürtekin-Demir 2002) among other sites. Our understanding Page 79 →of greater Lydia as a territorial entity and of its interactions with surrounding areas is gradually increasing (Roosevelt 2009). This period of expansion and consolidation of the Lydian state within western Anatolia also saw the emergence of Lydia as an important player on the interregional stage of the wider eastern Mediterranean. As well as their campaigns against the Ionian cities on the Aegean coast, it appears that the Lydians also struggled against the Cimmerians, fought with the Medes who had settled in central Anatolia, and conducted complex diplomatic relationships with the Neo-Assyrian, Urartian, and Egyptian royal houses. While Greek texts are mostly concerned with Lydian affairs within western Anatolia and especially on the Aegean coast, Neo-Assyrian documents are our best records of Lydian activity further afield. These tell us that a king known as Gugu (thought to be Herodotus's Gyges) petitioned Ashurbanipal for help against the Cimmerians and afterward entered into friendly relations sealed by the gift of Cimmerian prisoners of war to the Neo-Assyrian king. Later, however, Gyges seems to have fallen out of favor with the Assyrian king on account of his support of the newly emerging Saite dynasty in Egypt (Cogan and Tadmor 1977). Neo-Assyrian ambitions in Egypt had been curtailed by the rise of the Saites, and the sending of western Anatolian mercenaries to the aid of Psammetichus I in the 650s was seen by the Neo-Assyrians as an unprovoked act of aggression (Kuhrt 1995, 568-641). It is notable that these mercenaries included Lydians, Carians, and Greeks and that all three linguistic groups have left inscriptions and graffiti behind them in Egypt (Masson 1978). Diplomatic maneuverings such as these not only affected the Lydian elite at Sardis but also had an impact on communities across western Anatolia. An Egyptian stone statue now on display at the Hierapolis Museum, for example, appears to have been brought back home to the Upper Maeander area by one of Gyges's Carian mercenaries, returning home from the campaigns to aid the Saites in Egypt. It bears the following inscription in Greek: “Amphitheos's son Pedon brought me from Egypt and gave me as a votive.” The statue is said to have been found at Kale, stands about thirty centimeters tall, and is slightly broken at the shoulder. At the time of writing, it was yet to be formally published. As well as the expansion of Lydian power and the increased involvement of inland western Anatolian groups in the wider eastern Mediterranean, the seventh century also saw the floruit of the Greek cities on the Aegean coast. Despite the Greek textual tradition reporting regular offensives on the part of the Lydians, the growth of the Lydian state does not appear to have had any particularly adverse effect on the Greek cities' Page 80 →prosperity or cultural development. Indeed, the late seventh century is considered to be the start of the golden age of Ionia. During this period, the cities traded widely, establishing trading colonies in the Black Sea and central Mediterranean; produced high-quality artistic, literary, and philosophical work; and attained great wealth (Akurgal 1962: Boardman 1964; Crielaard 2009; Greaves 2010). The wide-ranging trade connections of the Ionian cities is particularly significant, not only for the riches it conferred on them, but also for the role it played in confirming the already-growing sense of East Greek identity. There is evidence that Anatolian Greeks from several cities banded into a single community when abroad, symbolically building a temple together, for example, at the Egyptian port of Naucratis (Herodotus 2.178). The close of the seventh century therefore sees western Anatolia peopled by a number of politically distinct and self-aware groups, including the various East Greek polities as well as the expanded Lydian state, which now covered much of inland western Anatolia. These groups interacted both

with each other and with groups beyond the region, on commercial, diplomatic, and military bases. In addition, Pedon's Egyptian adventures highlight the high levels of personal mobility on a regional and interregional scale during this time. With the opening of the sixth century, written historical sources become ever increasingly common, and the archaeological and epigraphic records are much fuller. The sixth century itself seems to have been a period of unparalleled wealth and prosperity in western Anatolia, with Lydia as well as the Ionian cities now entering into a golden age (both figuratively and literally) under the rule of the last Lydian king Croesus (Hanfmann 1978; Kuhrt 1995, 569-70). Twentieth-century commentators have remarked that “to be a citizen of Sardis at that time was to be a citizen of the most sophisticated country in the world” (Griffiths Pedley 1968, 4) and have described Sardis itself as “a Hollywood of opportunity and wealth” (Boardman 1999, 97). Even today, the city of Sardis itself is synonymous with legendary riches. “As rich as Croesus” runs the saying (Herodotus 1.32), and excavations at Sardis have gone some way to corroborating this image, uncovering evidence for gold refining and the minting of early forms of coinage (Davies 1994, 60-64; Ramage and Craddock 2000). The Greek cities on the Aegean coast also continued to thrive, leading the way in both economic and cultural innovation (Akurgal 1962, 375; Cobet et al. 2007; Greaves 2010). During the sixth century, the pre-Socratic philosophers Thales, Anaximander, and Anaximenes were active at Miletus (Waterfield 2000), as well as the geographer-historian Hecataeus (FGrHist 1; Pearson 1934), and monumental public building Page 81 →works were undertaken, such as the remodeling of the Heraion on Samos and the temple of Artemis at Ephesos (Osborne 1996, 262-64). Contacts between the Greek cities and the Lydian state seem to have been frequent and comparatively peaceful in relation to those of the seventh century. Both the Lydians and the Anatolian Greeks also seem to have interacted closely with the Greek groups of the central and western Aegean, the most famous example of which is probably the extravagant Lydian dedication at Delphi reported by Herodotus (1.50-52). The conquest of western Anatolia by Cyrus I in 547 BC did not immediately bring dramatic social and cultural revolution to the region but did signal the beginning of the region's historical role as a zone of political betweenness, caught between the ever more self-consciously Greek world and the Persian Empire. In addition, it ended the long period of autonomy enjoyed by the diverse communities of western Anatolia, making it an appropriate point at which to conclude this historical survey. The world of early sixth-century western Anatolia is thus no longer a world of protohistoric uncertainty and disciplinary betweenness. Rather, it belongs firmly to the spheres of classics and Achaemenid history, disciplines equally beset with their own types of much more historical uncertainties and betweennesses.

Overlooked Communities in the Upper Maeander Valley The history of LBA-IA western Anatolia is both rich and varied, and as already mentioned, many local areas show a considerable degree of autonomy for most of this period. One such area is the Upper Maeander river valley. The Upper Maeander is perhaps the least understood microregion within western Anatolia, receiving less scholarly interest than its counterparts on the coast. However, while western Anatolia as a whole is often considered to be caught in-between, the Upper Maeander valley is perhaps the most in-between area within the region. The Maeander river valley has always been and remains today one of the main channels of communication between the central Anatolian plateau and the Aegean coast (Marchese 1986). While there are several east-west routes and river valleys in the lower-lying areas closer to the coast, there are fewer routes further inland that lead the whole way up onto the plateau. The Upper Maeander river valley is therefore an important corridor for transport and communication (Thompson 2007; Thonemann 2006). It has rarely been Page 82 →considered as an interesting locality in its own right; attention has tended to focus instead on its connections to localities to its east and west. There is relatively little understanding of how people in the Upper Maeander area saw themselves—whether they thought of themselves as linked to the plateau, to the coast, or to neither. The communities here are perhaps the most overlooked in the already-overlooked region of western Anatolia. The Upper Maeander area is therefore a particularly appropriate setting in which to consider community identity. Both of the case study sites for this book, Aphrodisias and Beycesultan, lie in the Upper Maeander river valley. The two case study sites of Beycesultan and Aphrodisias are generally considered as belonging to a broad cultural

zone covering much of inland southwestern Anatolia (mentioned earlier in this chapter). This cultural zone was first identified in the extensive survey of southern Anatolia carried out by James Mellaart in 1951-52 (Lloyd 1954; Mellaart 1954, 1955). Mellaart claimed that stylistic patterning in his samples marked out a clear southwestern cultural zone, focusing around the inland Maeander valley area and constant throughout prehistory and protohistory. The southwestern zone appears to have had its own ceramic traditions and styles from the Chalcolithic period (Mellaart 1954, 188-89) through the Early Bronze Age (Mellaart 1954, 196-207), the Middle Bronze Age (hereafter MBA; Lloyd and Mellaart 1965, 77), the LBA (Lloyd 1954, 217; Mellaart and Murray 1995, 99-109), and the IA (Mellaart 1955, 115, 130). Since then, the existence of a general area of cultural similarity in inland southwestern Anatolia has been borne out by both survey and excavation (Abay and Dedeo lu 2005, 2007; im ek 2007; Thompson 2007), and the two main sites that have been excavated in this area to date—Beycesultan and Aphrodisias—show a notable similarity in their ceramic styles and development (Joukowsky 1986, 429, 460-69). It is hoped that new excavations at the site of Laodicea will shed more light on this cultural zone in years to come ( im ek 2007). In general, cultural zoning should primarily be considered as an analytical tool, rather than evidence for meaningful social entities over the landscape. This is now widely accepted in archaeology and has been explored with reference to social identities such as ethnicity, which are no longer thought to map directly onto cultural patterning (Jones 1997; Gamble 2008, 187ff.; Mac Sweeney 2009). The inland southwestern zone of Anatolia is no exception, and there are indeed specific reasons why this area may not correlate with any social phenomenon. The ceramic assemblages of Upper Maeander sites are characterized by a notable degree of Page 83 →similarity but not uniformity. In addition, it should be recognized that the southwestern Anatolian cultural zone relates almost exclusively to pottery styles and does not appear to have manifested itself in other cultural similarities, such as mortuary forms or architectural techniques. It has sometimes been thought that crescent-shaped loomweights are also characteristic of the southwestern zone, suggesting some similarities in textile technologies and patterns of production (Mellaart and Murray 1995, 118). However, there is some controversy over whether these objects should be considered to be loomweights (Vogelsang-Eastwood 1990), and examples of these objects have been found elsewhere in Anatolia and in especially large numbers at Karahüyök in the Konya plain (Alp 1968) and Demircihüyök in central Anatolia (Kull 1988). They cannot, therefore, be said to be characteristic cultural features of the southwestern zone. In contrast to this, the Aegean coast and island zone can be seen to have shared some similarities in social practices, such as mortuary ritual in the LBA and linguistic, architectural, cultic similarities in the IA, evolving into a shared social and political identity in the later IA (see previous sections of this chapter). The inland southwestern zone should therefore be viewed as a useful analytical category, rather than a meaningful social entity. As already mentioned, the category is based primarily on ceramics, and the ceramic assemblages of all excavated sites in the Upper Maeander region bear notable similarities. Ceramic typologies have been developed independently for the two main excavated sites, Aphrodisias and Beycesultan, and are aimed primarily at establishing chronologies for the sites (Joukowsky 1986; Mellaart and Murray 1995). Both ceramic assemblages have also been studied for evidence of external contacts in the form of imports and stylistic influences (Joukowsky 1986, 427-76, for Aphrodisias; Mellaart 1970 and Mellaart and Murray 1995, 99-109, for Beycesultan). However, there is no unified system of classification that can be applied to the pottery of both sites to allow for a detailed comparison between them. The Upper Maeander valley would therefore benefit from a coherent ceramic typology. The outline of such a classification scheme is described in the following section of this chapter. This typology allows for the comparison of ceramics from the two case study sites that forms part of the analysis in chapters 7 and 8, and it has been used in the catalogs of ceramic vessels in appendixes B and D. The appendixes present a complete catalog of all ceramics and small finds known from both case study sites, offering data that has until now only been partially published. The following typology Page 84 →has been a critical element that has allowed this comprehensive cataloging. As presented here, this classification scheme is not meant to provide an accurate means of distinguishing chronological variation within either of the case study sites, nor is it meant to present a unifying pottery sequence for the southwestern Anatolian cultural zone as a whole. It is simply a tool by which the ceramic

assemblages of LBA-IA Aphrodisias and Beycesultan can be compared, establishing a common vocabulary between the two. In the following section, a scheme will be outlined for considering Upper Maeander ceramics according to, first, decorative type and, second, vessel shape.

Upper Maeander Ceramics Ceramics by Decorative Type Ceramics from the Upper Maeander valley can largely be classified into the following broad decorative categories. These categories have been drawn up not on the basis of strict formal elements but, rather, according to the type of external associations they were likely to have evoked for communities in this area. As already discussed in chapter 5, styles of material culture can recall associations of the external Other if they simply evoke a “foreign” style. They do not necessarily have to conform to the official stylistic criteria but can merely give off enough visual cues to create a connection in the mind of the viewer with that external Other. The categorization of ceramic decoration that follows takes this into account and aims to distinguish decorative schemes that would have evoked specific external Others for the people of Aphrodisias and Beycesultan. A. Lustrous

A1. Silver or gold lustrous. These finewares are coated with a micaceous wash, sometimes but not always polished, to produce a shimmering surface. Colors include gold, silver, and a gold-silver mix, under a micaceous sheen on the surface, making for a metallic finish overall. Vessels with such decoration are mainly associated with the area around the Hermos river valley. At sites such as Bayrakli and Sardis on the river Hermos, this style has a long history, stretching back into the MBA (Mellaart and Murray 1995, 105-6). Mellaart's survey suggests that the LBA distribution of lustrous wares was concentrated in the Hermos valley in particular, although, Page 85 →at this time, they were also found in smaller quantities in the Troad and the Maeander river valley (Mellaart and Murray 1995, 130). The relatively late arrival of this general style in the Upper Maeander area, however, makes it likely that during the LBA at least, the broad aesthetic of gold or silver lustrous washing would have been conceptually associated with areas to the northwest and the Hermos river valley. A2. Red or coppery lustrous. Also a fineware style, these vessels were finished with a similar type of micaceous wash to that used for Type A1. However, the final colors vary between a light coppery red to a darker brown red, producing a very different overall visual effect. Vessels decorated in this manner have a very different spatial and temporal distribution than those belonging to the previous category. In terms of geographic distribution, they are concentrated in the area around Beycesultan and the Çivril valley (Mellaart and Murray 1995, 130). In terms of dating, they first appear at the end of the MBA, much later than the gold and silver lustrous wares, although both types only become popular to any extent in the LBA (Mellaart and Murray 1995, 106-7). This style should therefore be broadly considered as local to the Beycesultan and Çivril valley area and as recognizable as such, although it incorporates the aesthetic of the lustrous wash from the Type A1 styles of the Hermos valley region. B. Linear Painted

B1. Linear painted decoration on a lustrous surface. This style is strongly associated with Aphrodisias, as it appears there in substantial quantities in the first two protohistoric phases of occupation (table 7). The silver and gold lustrous washes of Type A1 are used as a base for linear painted decoration, often in red or orange paint. The painting is tight, in dense patterns with designs laid out relatively closely to each other. The style is only found at a small number of locations beyond Aphrodisias and only ever in small quantities, and it should therefore be considered as local to the Dandalus valley area. It should be thought of as a local Dandalus valley innovation on the Hermos valley A1 style, in the same way that Type A2 should be seen as a Beycesultan/Çivril valley innovation on the A1 aesthetic. B2. Linear painted decoration on a matt surface. This is a broad category, incorporating all vessels that have linear painted decoration on a matt surface. The breadth of this category means that it does not represent any

specific Page 86 →styles or particular geographic connotations, and the visual cues given by vessels in this category must be considered on an individual basis.

C. Warm-Colored

C1. Warm-colored, burnished. Warm colors include reds, browns, and oranges, and these colors are strongly represented in the ceramic assemblages of both case study sites. This category and the following category, Type C2, refer not to warm colors resulting from the natural hue of the clay but to vessels where a deliberate decorative effort has been made to produce one of these colors. Pieces of Type C1 are usually slipped or washed and then burnished to produce the sheen characteristic of the style. The category also includes warm colored vessels that have been pattern burnished, the most common of which is radial burnishing on the inside of open vessels such as bowls, kraters, and chalices. This style is highly distinctive and is again closely associated with the Upper Maeander area, in particular with Beycesultan and the Çivril valley during the LBA (Mellaart and Murray 1995, 1-2, 107-8). C2. Warm-colored, slipped. This category is another general one, which includes all vessels mostly showing warm colors—reds, browns, and oranges—but that are not highly burnished. Like vessels of Type B2, they are not necessarily associated with any specific geographic area or ceramic style, although vessels within this category will be considered individually for the specific visual cues they may present. D. Grey and Black

The visual impact of grey or black surfaces is quite different from that of surfaces in warm colors (Type C1 and C2). This category includes all vessels where the dominant surface color is grey or black, whether the finish is slipped or burnished. There is no tradition of dark-colored surfaces on ceramics in southwestern Anatolia, but grey wares in particular are thought to be characteristic of northwestern Anatolia throughout the Bronze Age (Bayne 2000) and into the EIA (Chabot Alsan 2009). It is therefore proposed that vessels with a black or grey finish would have recalled associations with the northwest for the case study communities, especially the far northwest, including the Troad and the north Aegean islands. Page 87 → E. White and Pale-Colored

This type refers to vessels that have predominantly pale or white surfaces, sometimes due to slipping, but also potentially due to painting. Pots with painted decoration on top of this pale surface are also included in this category rather than in Type B2, as long as the decoration remains of secondary visual impact to the pale colors of the surface overall. The White Painted style and white slipped pottery more generally are well known from southeastern Anatolia around Tarsus and Mersin (Artzy 2001), the Amuq, Cyprus, and the Levant, where they are most frequently associated with the EIA (Karageorghis 2001). In addition, dark-on-light painted decoration is characteristic of the EIA Levels 2e-2j at Kilise Tepe (Postgate and Thomas 2008, 123-24). While few of the vessels included in this category could properly be considered as belonging to any of these “official” whitebackground styles, the aesthetic of a pale surface is, in itself, distinctive and unusual. This type therefore includes all vessels that might have evoked white styles, although not conforming to any strict stylistic standards. I would suggest that such vessels would have had connotations of the southeast at Aphrodisias and Beycesultan. F. Plastic Decoration or Incision, No Other Surface Treatment

Again something of a catchall category, this includes appliqué decoration, stamping, knobbed decoration, incised designs, and “rope decoration.” As with other general categories, individual vessels will be considered separately. G. Known Ceramic Styles

G1. Mycenaean. Mycenaean styles of ceramics have been extensively researched and are now understood not only as an overall style but in their regional variants as well (Mountjoy 1993, 1999). Certain types of Mycenaean pottery, however, have an extremely wide distribution across the eastern Mediterranean because they were widely traded in the LBA long-distance networks of exchange that have been the subject of such intense investigation (as discussed earlier in this chapter). Not all Mycenaean pottery, therefore, would have necessarily held specifically Aegean connotations for the communities at the case study sites. The sole example of a Page 88 →Mycenaean vessel found in the case study sites is at Beycesultan (Cat. no. B1.114), and since this belonged to a stirrup jar, one of the most widely traded ceramic items of the LBA, it is unclear whether it would have recalled Aegean connections at all. Instead, it may have been more generally reminiscent of the wide-ranging interregional trade networks of the LBA. G2. Phrygian. The Phrygian vessels from Beycesultan were never cataloged or properly documented, although some drawings have been published (see chapter 7). They were not kept after the excavations and so could not be analyzed firsthand, but they still comprise a readily recognizable style of pottery that would have likely evoked external associations for the community at Beycesultan. Phrygian styles of ceramics are also already documented, and some typologies and chronologies are available (Johnston 1970; Sams 1974, 1994). These vessels would have likely recalled links with Phrygia on the central Anatolian plateau to the northeast, especially since they did not recall more generalized external associations, as they were not widely traded in the way that some Mycenaean shapes were. G3. Lydian. Like the Phrygian pots from Beycesultan, the Lydian vessels from Aphrodisias belong to a readily identifiable group and a documented style. Lydian ceramics are already well known and recorded from sites such as Sardis (Greenewalt 1966, 1968). In addition, the vessels from Aphrodisias have themselves been comprehensively analyzed and published (Mierse 1986). Like Phrygian ceramics, Lydian pottery was never widely traded, so it would have probably carried specific connotations of the Hermos valley area and possibly even of the Lydian state. G4. East Greek. East Greek pottery—from Ionian Geometric to the Wild Goat style, from bird bowls to Chian Grand style—has been well researched, and classification schemes have been drawn up with a high level of both geographical and chronological accuracy (Cook and Dupont 1998; Coldstream 1968). Like Mycenaean pottery, some styles and shapes of East Greek pottery were widely traded around the eastern Mediterranean, especially after the settlement of Greek colonies at Tarsus in southeastern Anatolia and Naucratis in Egypt. The associations provoked by an East Greek vessel would therefore depend on the particular vessel, and each vessel will again be dealt with on an individual basis. Page 89 → H. Undecorated

This category includes all vessels with no notable surface treatment. X. Unknown

Due to the way ceramics were documented at both of the case study sites, some vessels were recorded in a very brief fashion, and there are several cases in which no reference is made as to their decoration or style. While this may have been because these pots were undecorated, we cannot be sure of this. When there has been no record of a vessel's decoration and when the vessel has not been available for firsthand analysis, the vessel in question has been placed in this category. Ceramics by Vessel Shape: Functional Category and Characteristic Shapes In the chapters that follow and in the catalogs of vessels contained in the appendixes, each vessel has been classified not just by its decorative type but also by shape. These shapes are important for the analysis on two counts: first, in terms of their implication for the functions of the vessels; and second, in terms of any external

associations they may have evoked. The functional category will help point toward whether the vessels were used as part of enactments of community or not. While this can only be fully ascertained by considering the find context of the vessels, the shape also constrains their functional uses. The functional categories and shapes designated here are deliberately general and broadly fall into line with standard ceramic schemes established elsewhere. They also derive heavily from Mellaart's typology for Beycesultan, although references are given for the definition of each vessel shape. In addition to information concerning the use of the vessels, the shapes can also imply connections with external groups. Certain shapes may be closely associated with particular geographical areas, and it is possible that they would have been recognized as conspicuously foreign or local in the Upper Maeander. Where this is the case, the likely association is noted as part of the description of the vessel shape. Where no specific geographic association is noted, it is assumed that the shape cannot be firmly linked to any particular area. Page 90 → Dining

Dining, especially communal dining, is often invested with social significance and, as already discussed, is a likely arena for the creation and contestation of social identities. This intentionally broad category includes vessel shapes usually associated with individual service, communal service, individual drinking, and pouring. The inclusion of several different subcategories under this one attempts to reflect the likely social context of these vessels' use, which is the issue of central interest in this study. General Service 1. Bowl: general category for mostly simple bowls with straight profiles. 2. Bowl, carinated: bowl with sharply carinated profile. 3. Bowl, Ionian: bowl of East Greek shape; used here as a generic term for bird, eye, and rosette bowls; characteristic of the Aegean coast. 4. Bowl, spouted: bowl with spout attachment. 5. Dish: plate or shallow bowl with everted rim or flat profile. This approximates to Rice's definition of “plate” as a vessel for which height is less than one-fifth of the maximum diameter (1987, 216). Communal Service 6. Bowl, deep: large bowl with deep basin, of any profile, intended for communal service. 7. Fruitstand: large bowl on long stem, intended for communal service (Mellaart and Murray 1995, 3); characteristic of the Upper Maeander area. Drinking 8. Chalice: stemmed cup with sharply carinated profile (Mellaart and Murray 1995, 3); characteristic of the Upper Maeander area. 9. Cup: general category for small individual drinking vessels, usually with rounded bottom and one handle. 10. Cup, conical: cup with conical profile. 11. Cup, carinated conical: cup with carinated conical profile, usually with a foot; characteristic of the Aegean. 12. Cup, conical-footed: round cup with conical foot; characteristic Page 91 →of Lydia and the Hermos valley

region. 13. Cup, bifoil: cup with bifoil mouth. 14. Cup, quatrefoil: cup with quatrefoil mouth; characteristic of the Upper Maeander area. 15. Mug: drinking vessel with straight vertical sides; characteristic of the Upper Maeander area. Pouring 16. Jug: general category for pouring vessels. 17. Jug, beaked: jug with beaked spout; characteristic of central Anatolia. 18. Jug, cutaway spout: jug with cutaway spout. 19. Jug, round mouth: jug with round mouth. 20. Jug, bifoil: jug with bifoil mouth. 21. Jug, trefoil: jug with trefoil mouth. 22. Jug, quatrefoil: jug with quatrefoil mouth. 23. Lydion: jug with Lydian shape (Mierse 1986, 413); characteristic of Lydia and the Hermos valley region. 24. Teapot: squat pouring vessel with thin spout and high basket handle (Mellaart and Murray 1995, 5); characteristic of the Upper Maeander area. Storage

This category includes both vessels that would have been used for large-scale agricultural storage (which would have been less likely to have been used in social signaling) and also vessels for small quantities of valuable substances. Again, reference must be made to the individual vessels when determining social use. 1. Jar: general category for closed vessels. 2. Jar, small: small jar. 3. Lentoid flask: vertical flask made from two hemispheres and with long neck, sometimes called a pilgrim flask (Blegen, Caskey, and Rawson 1953, 60 [Blegen shape B41]); characteristic of central Anatolia and the Levant. 4. Stirrup jar: Mycenaean small jar with “stirrup” handle and spout (Mountjoy 1993, 40-42); characteristic of the Aegean. 5. Askos: horizontal flask, often with three handles, sometimes on a foot; various forms. Page 92 → 6. Pithos: large storage vessel with open neck. 7. Jar, storage: large storage vessel with closed or semiclosed neck. Cooking

These vessels are usually undecorated and made out of coarse ware and are therefore thought to have been used in

food preparation. Several of these vessels show signs of blackening on the outside from being placed over fires. 1. Cooking pot: general category including many blackened pots. 2. Colander: coarse dish with holes in bowl. 3. Drink warmer: shallow bowl over chamber for a heat source (to keep contents of the bowl warm), mounted on pedestal base (Mellaart and Murray 1995, 25); characteristic of the Upper Maeander area. 4. Potstand: round stand, often blackened, on which cooking pots could have rested (Mellaart and Murray 1995, 5). 5. Tripod stand: tripod stand, of similar function to the circular potstands. Ritual

While many vessel shapes may have been used in ritual, only these three shapes, representing five vessels in total, can be unambiguously defined as “ritual.” 1. Rhyton: long vessel open at both ends, for liquid to run through. Libation arms and funnels have been included within this category. Libation arms are characteristic of central Anatolia, especially during the LBA. Funnels are more typical of the Upper Maeander area. 2. Urn: there is only one burial known for certain from the case study sites, the cremation of the later IA period from Aphrodisias, associated with a mortuary jar referred to as the “Carian Jar” (Benjamin 1986). 3. Dipper: or bucket. Page 93 → Unclear 1. Sherd: body sherds from which it is impossible to guess at a complete profile or even the general shape of the vessel. 2. Handle: although it is possible to identify some handles as belonging to pouring, storage, or cooking vessels, this is not always possible. Handles for which uncertainty remains have been included in this category. Communities, In-betweenness, and the Case Studies Western Anatolia is an often overlooked region, dismissed as a border zone, a contested periphery, and a place inbetween. Within western Anatolia, the Upper Maeander area has been especially characterized as “between,” due to its location on a major channel of communication between the central Anatolian plateau and the Aegean coast. As a result, the communities of the Upper Maeander area have rarely been considered in their own right, and their community identity has been especially overlooked. The Upper Maeander area is therefore an ideal setting in which to investigate the dynamics of community identity. In the two case studies that follow, evidence for community identity will be considered and interpreted, highlighting the unique developments of different communities at different settlements. The case study sites themselves, Aphrodisias and Beycesultan, are very different on several counts: in terms of their location within the Upper Maeander area, in terms of their size and central focus, and also in terms of the material available for study from each. In addition, each settlement develops along its own independent trajectory during the LBA and IA, reacting differently to the same historical conditions and events. However, there are enough similarities between the two sites to allow for comparison between them, with pottery in particular offering good opportunities for contrasts and comparisons. To enable such comparisons to be made efficiently, a

coherent ceramic typology has been outlined in the previous section. Although this typology is not designed to aid chronological investigations, it should nonetheless facilitate the comparison of broad social developments at the two sites, particularly with reference to the performative representation of external Others.

Page 94 → CHAPTER SEVEN

Communities under Pressure at Beycesultan The Site and the Material Introduction With the exception of Troy, Beycesultan is perhaps the most widely known and frequently discussed site of the western Anatolian Bronze Age (fig. 4). At the time of its main excavation in the 1950s, it was showcased in the popular press as the capital of an important “forgotten nation,” on account of the impressive nature of the remains uncovered (Lloyd 1955). Today, it is considered to be the “type site” for inland southwestern Anatolia, in much the same way as Troy is considered to be the type site for the northwest (Bryce 2006, 45, 47, 76; Mellaart and Murray 1995, 108). New excavations were initiated in 2007, with the aim of clarifying some of the unanswered questions left from the original explorations of the mound (Abay and Dedeoglu 2009). Hopefully this new research will bring novel insights to our understanding both of the site and of the region more generally. However, as data from the new excavations is still under initial analysis, this case study focuses on material from the original excavations in the 1950s, which is in much need of reinterpretation. It is hoped that this reinterpretation will provide a solid foundation for the insights that will undoubtedly come from the recent excavations in the following years. In addition to this recent archaeological work on the site, attempts have also been made to discover the site's specific identity in contemporary Hittite documents, and while it is likely that Beycesultan is to be located in the Arzawan territory known as Kuwaliya, this remains unconfirmed (see chapter 6 and discussion in the present chapter). The site itself lies about five kilometers southwest of the town of Çivril in the Denizli province ( fig. 3), at the northeast end of the fertile intermontane valley of the Upper Maeander river. The valley is ringed with low hills and located on a natural route from western Anatolia to the Anatolian plateau. Although the modern road to central Anatolia now runs through the more southerly valley of Dinar, it seems that the main route passed through the Çivril valley at several points in history (Thonemann 2006). At these times, the road would have passed within a few kilometers of Beycesultan, climbing onto the plateau through a pass between the modern villages Çivril and I ikli. This pass would have provided access not only to central Anatolia but also directly north toward the Gediz river valley and the region of modern U ak. At the opposite end of the valley from Beycesultan, to the southwest, the road leads down to Denizli and eventually to the lower Maeander plain, abruptly dropping about two thousand feet through a steep mountain pass. Page 95 → Unexplored höyük mounds litter both this valley and the lower Denizli plain, indicating the dense occupation of the landscape in antiquity (Abay and Dedeoglu 2005, 2007; Mellaart 1954, 1955). But with its twin mounds rising between 18 and 25 meters above the plain and stretching 850 meters across at their greatest extent (fig. 4), Beycesultan remains one of the largest and most prominently visible höyük sites in the Denizli province and is the most intensively investigated höyük in this complex archaeological landscape to date. The remains of Beycesultan comprise two mounds and a sherd scatter over a substantial area around their base, suggesting that the settlement extended over several hectares. Intensive systematic survey is still needed, Fig. 4. Beycesultan site plan. (Reproduced with the kind permission of the British Institute of Archaeology at Ankara.) however, to gauge the extent of occupation during different periods. Most of the excavation was carried out on the east mound (fig. 4). A few trenches and a long section were opened in the west mound, and several exploratory trenches were dug around the bases. Occupation seems to have been fairly continuous from the Late Chalcolithic period right up until the IA, with forty archaeological levels identified and labeled by the excavators (Levels I-XL) covering this time. The last four of these—Levels III, II, Ib, and Ia—are relevant to this study, as they span the LBA and the IA. Page 96 →

After these formally described archaeological levels, there is evidence for the site's occupation at different points in its later history, in particular during the Phrygian period. However, little is known about these occupation levels. Indeed, the “Phrygian” houses and ceramics of Trench Z have a single sentence devoted to them in the initial excavation report (Lloyd and Mellaart 1962, 12) and only four pages (only one of which is text) in the official site publication (Mellaart and Murray 1995, 88–91). There seems to have been an extensive but almost completely undocumented Byzantine settlement and cemetery as well. The Byzantine remains are discussed in a single paragraph in each of the excavation reports (Lloyd and Mellaart 1962, 9; Lloyd and Mellaart 1955, 40). The Page 97 →uncertainty about the later IA, classical, and Roman periods at the site is due, however, not just to the lack of interest of the excavators but also largely to activity on the site in the Byzantine period. During this time, the west mound was artificially leveled to create a flat emplacement for the settlement, and the east mound was used as a Christian cemetery (Lloyd and Mellaart 1962, 9). The latest known use of the site was the construction on the east summit of the Ottoman tomb that gives the site its name—Behicesultan. The tomb is still in good condition and is currently maintained as a local shrine. The site's long history, its prominence in the archaeological landscape, and its centrality to our understanding of the western Anatolian past all make it an intriguing case study for this book. Problems and Solutions Despite its relative fame and importance, the site of Beycesultan is still not well understood and presents several problems to the analyst. During the original excavations, there were six seasons of excavation in total between 1954 and 1959 under the direction of Seton Lloyd and James Mellaart, allowing for a maximum of 295 days of excavation.1 During this time, over five thousand square meters—a large area by modern standards—was uncovered to a maximum depth of twenty-eight meters (Mellaart 1999, 61). The rapid pace of the work and the impressive breadth of the project's scope led to some inevitable compromises in the level of detail in the recording. Significantly, all the material considered in this case study was excavated in just two years, with the stated aim of the 1955 excavations being the speedy clearance of Levels I-IV in order to uncover the MBA levels underneath (Lloyd and Mellaart 1956, 101). Lloyd discusses a few choice finds from the LBA-IA levels as being compensation for the “somewhat tedious work of clearing the upper levels, in preparation for a return to our main objective beneath” (Lloyd 1955). In addition, policies for the documentation of finds, while in line with the common practice of the time, preclude methods of modern quantitative analysis, as only items chosen for publication or sent to museums were recorded. For ceramics, this meant finewares and “unusual” shapes of coarse ware, in mostly complete or near-complete states of preservation. Among the small finds, for frequently occurring objects such as spindle whorls and loomweights, only a selection thought to be representative were recorded, or examples that deviated from the norm (Lisa and David French, personal communication). Any quantitative analysis of this material, therefore, must be treated with extreme caution. The numerical quantities and percentages presented in Page 98 →this case study are given only as a means of rough comparison, without any claim to be fully representative of the excavated assemblage. Fortunately, the excavation and recording biases are such that the material can still be meaningfully interpreted. First, the biases are constant throughout the different levels of excavation. This means that relative proportions can be compared from phase to phase, even if the absolute numbers cannot be relied on. Second, the nature of the biases is known, as are the clear selection patterns. By choosing specifically to document imported items and the more unusual, eye-catching, and attractive finds, the excavators will have included many of the objects most likely to have been invested with some social significance. In addition, as the documented finds were also selected specifically in order to show the full range of material found, we should still have some concept of this range of artifacts, even if we have no accurate data about the quantities of objects within this range. Therefore, although it may not be representative of the complete archaeological assemblage, the recorded material still has a lot that it can potentially tell us about community identity at LBA-IA Beycesultan. Chronology The dating of Beycesultan's LBA and IA levels is problematic, as no samples were taken for scientific dating during the original excavations. In addition, the ceramic assemblage is highly idiosyncratic, with very little

evidence of imports or external influences on either decorative styles or shapes to tie the Beycesultan sequence in with the more established pottery chronologies from elsewhere in the eastern Mediterranean (Mellaart and Murray 1995, 99). For decades after the original excavations, there was much uncertainty about how the various levels should be dated (Mellaart 1970; Mellaart and Murray 1995, 94–96). Carbon dating undertaken as part of the new excavations promises to resolve this uncertainty in the future. Until these results become available however, there are a small number of ceramic finds that help to locate the Beycesultan levels in relation to developments elsewhere. The chronology argued for here is a relative one, and the absolute dates that are generated for the levels at Beycesultan are dependent on the absolute dates assigned to better-established pottery sequences. These are, of course, themselves the subjects of debate, and it is likely that the absolute dating of Beycesultan will change in the future. The first peg on which the Beycesultan LBA-IA chronology can be hung belongs to the very start of the LBA period in Level III and comprises a single sherd from a Mycenaean stirrup jar built into a platform in Page 99 →the Trench J extension (fig. 7; Cat. no. B1.114; the sherd is also mentioned in Mellaart and Murray 1995, 2, 93). Although a solitary find, this gives us a terminus post quem for the level's construction, placing it no earlier than 1360 BC. Mellaart suggests, while admitting that this is “not precise enough to offer any chronological comfort,” that Level III should therefore be dated broadly to the fourteenth century (Mellaart and Murray 1995, 96). However, given the potential for curation of such an item as a stirrup jar and given the time likely taken between the vessel's production and its final deposition in the platform, I consider it unlikely that the final deposition of the sherd would have been close to the earliest date of possible production in 1360. I therefore concur with Lisa French's rough dating of the level's initial construction to +/-1300 BC (Lisa French, personal communication). Given that the level was partially rebuilt and repaired during its occupation in a subphase that Lloyd calls Level IIIa (Lloyd 1972, 9), it seems that Level III as a whole should be considered as continuing for a considerable length of time, and I therefore think that Level III began roughly around the turn of the thirteenth century and that its principal period of occupation must have been several decades spanning the early to middle thirteenth century. The next point at which Beycesultan can be linked to better-known chronologies is not until the end of Level II, when a small number of vessels appear (including flasks and bottles), which correspond directly to central Anatolian coarse wares. While these vessels were unfortunately never fully described, illustrated, or documented, Mellaart reports that they belong to a type known from the closing years of the Hittite Empire and that continued to be used even after the collapse of the Hittite state in ca. 1180 BC (Mellaart and Murray 1995, 93). This suggests that the final phases of Beycesultan Level II were either exactly contemporary with or, more likely, slightly later than the fall of the Hittite state and belong to the mid-twelfth century. In this, I concur with Mellaart's assertion that Level II should therefore be viewed as dating from the thirteenth to the mid-twelfth centuries. The emergence of linear painted wares in any notable quantity is documented in Level Ib (Types B1 and B2 in table 4). The quantity of these sherds and their fabrics suggest local production (Mellaart and Murray 1995, 94), and their advent at Beycesultan is paralleled by their appearance across western, central, and southern Anatolia in the IA (Cook 1958–59, 10; Hansen and Postgate 1999, 111; Postgate and Thomas 2008, 123–24; Spier 1983, 23). However, as IA chronologies are still uncertain all over Anatolia and the eastern Mediterranean more generally, firm dates still cannot be assigned to Level Ib Beycesultan. However, these simple Page 100 →linear forms are generically thought to belong to the late twelfth, eleventh, and tenth centuries, spanning the Submycenaean and Protogeometric periods on the Aegean coast and the Early Phrygian period in central Anatolia. Level Ib must therefore have been constructed in the middle to late twelfth century, after the closing stages of Level II, and likely continued into the eleventh, while Level Ia must have followed it some time during the eleventh century. The vagueness in the dating of these two periods is unfortunate but reflects the vagueness surrounding these centuries in Anatolia as a whole. As mentioned earlier, the uncertainty about whether or not and when occupation at Beycesultan was discontinued is a product of the later Byzantine occupation of the site. The final recognizable IA occupation level is the Phrygian phase, which can confidently be dated to the eighth century by the recognizable Phrygian finewares (Mellaart and Murray 1995, 89). It is expected that firmer absolute dates will become available for Beycesultan's LBA and IA levels in the near future. However, the current imprecision in dating is perhaps not as problematic as it might initially seem. While

precise dates in terms of years cannot be assigned to any of these levels, there is always a sense of relativity to contemporary sites, allowing comparisons to be made between broadly contemporary developments at Beycesultan and elsewhere. The subject of this study is this comparability and the interrelation of different communities, rather than the determination of exact temporal boundaries. Despite difficulties with its chronology and the surviving records and archaeological material, Beycesultan is still of great importance to prehistory and protohistory as one of the few excavated settlements in inland western Anatolia. The problems associated with it are themselves not insurmountable, and the material remains a worthwhile subject of research. The following section outlines the practicalities of exactly what material was available for consideration in this study and how it was accessed. Practicalities The material considered in this case study comes mainly from finds published in the most recent excavation monograph (Mellaart and Murray 1995) but also relies on a number of other sources of information for supplementation. The archives of the British Institute of Archaeology at Ankara (BIAA) contain valuable documentary records of the excavations, including field diaries, plans and photographs, and the finds register book. The register book details both the ceramics and the small finds accessionedPage 101 → to the Turkish authorities by the excavators. After their registration, these objects were distributed among museums in Ankara, Izmir, and Denizli, with the bulk of material ending up in the Hierapolis depots of the Denizli provincial museum. The information the register book offers is variable but nonetheless an important supplement to the material published by the original excavators. Both are presented in this book. For small finds, information from the register was used in addition to Ann Murray's excavation catalog (in Mellaart and Murray 1995), which is itself based on both the register book and a detailed study of the small finds still stored in the Hierapolis Museum depots. Murray's catalog includes 171 small finds that can be assigned with certainty to the LBA and IA (Levels Ia-III), all of which are considered here. However, 59 additional objects from these levels were recorded in the register book but not included in Murray's catalog, as they could not be located in the museum stores at the time of the study. Therefore, the total number of small finds considered here is 230. A summary of the small finds considered in this case study can be found in table 1, and fuller records of them are in the catalog in appendix A. Published ceramic vessels from the LBA and IA levels number 470. In addition to these, the register book contains 107 vessels that were registered as belonging to one of these levels but never published, bringing the total ceramic sample for this study up to 551 vessels. In the absence of a recent study similar to Murray's small finds catalog, the information from both the register book and the publications is supplemented by the author's own work on the pottery in storage at the Hierapolis depots.2 While 160 vessels from relevant levels were accessioned by the Turkish government, only 112 were found in the museum stores, while 48 could not be located. Tables 2, 3, and 4 give a summary of the ceramic vessels considered in this case study, while full records of these vessels can be found in the catalog in appendix B. Unregistered sherds from Beycesultan are to be found in a number of museum and university collections, including the Museum of Archaeology and Anthropology in Cambridge, the BIAA collections, and the University College London sherd collection. I have decided not to include these sherds in this study because they were not subject to the same consistent selection processes as the finds selected for official registration but were chosen in a more haphazard way. The catalogs presented in appendixes A and B comprise, for the first time, a complete inventory of all objects and ceramics from Beycesultan for which we have any record. While some of these items have been published previously, many others have come to light both in the excavation stores and through the surviving documentation. Appendixes A and B Page 102 →therefore represent the most complete record currently available for both finds and ceramics from the 1950s Beycesultan excavations. Although assembled from various different sources and unrepresentative of the complete excavated assemblage, this body of material is still a rich and valuable source of information.

Level III

The Remains of Level III The construction of Level III (ca. 1300 BC) followed directly from the preceding MBA Level IV (fig. 5), and the remains suggest a time of general socioeconomic renewal at Beycesultan, including a “renaissance” in the ceramic traditions (Mellaart and Murray 1995, 1–2) and major construction work undertaken in all excavated parts of the site (Lloyd 1972, 4). In addition, there was a return to centralized urban planning that had not been seen since Level V. The large area uncovered on the east mound shows domestic units ranged along two straight, parallel roads, as well as traces of a stone settlement wall around the mound's edge. In addition to these signs of central planning in the domestic areas, the twin shrine buildings in Area R to the north of the mounds were rebuilt in Level III along similar lines to their MBA predecessors (Lloyd 1972). Two factors are relevant in this remodeling of the site, and both suggest a high degree of social stability and complexity at Beycesultan. First, the social capital needed both to undertake and regulate these construction projects is substantial and indicates some structures of status and authority. Second, the continuity of cult practice in terms of both location and style of cult installations and forms of ritual activity is a significant marker of symbolic continuity with the past and a conscious stress on social stability (Lloyd 1972, 34–36; Lloyd and Mellaart 1958). This sense of tradition and stability is reinforced by the way the structures were rebuilt in the subphase Level IIIa—they were rebuilt on exactly the same lines as before, with no significant deviation (Lloyd 1972, 9). In addition, there may have been a central building on the west mound at this time, as a rectangular building that appears to have pillared supports and dressed masonry was excavated there. Unfortunately however, its exact function remains uncertain (Lloyd 1972, 21). At the same time, the use of sealing practices suggests a level of complexity and organized administration at the site (cf. Cat. no. B1-60, stamp seal), and robustness in Beycesultan's economy is indicated by the presence of several items probably associated with prestige and status. These include ornamented daggers (Cat. nos. B1-01, B1-02, B1–44 to B1–50), a highly decorated axhead (Cat. no. B1–04), a bowl made from serpentine (Cat. no. B1–52), and pieces of decorated horse harness (Cat. nos. B1–63 to B1–66), as well as objects made from highvalue materials such as rock crystal (Cat. no. B1–55), carnelian (Cat. no. B1–56), faience (Cat. no. B1–70), and silver (Cat. nos. B1–37 to B1–39).

Prosperity and interest in status objects are also suggested by the innovations in ceramics (Mellaart and Murray 1995, 1–2). Mellaart remarks that compared to the previous MBA Level IV, the Level III assemblage contained a larger quantity and greater range of finewares, the changes in style and shape being accompanied by an overall higher quality of technical workmanship. The fineware assemblage has a very particular aesthetic, involving metal skeuomorphism in shapes and fine pattern burnishing of red to brown slips in decoration. A rough categorization of vessels registered or published from Beycesultan into decorative types (table 4) shows that of all levels, Level III shows the greatest standardization in the ware types of its ceramic assemblage. It has the highest proportion of its recorded ceramics belonging to a single type, with 56.7 percent of its ceramics belonging to Type C1, warmcolored burnished wares. Taken together with Type C2, warm-colored slipped wares, this means that 75.6 percent of recorded vessels had a very similar decorative scheme and would have presented a very similar visual image. In addition to this, analysis of vessel shape also shows a preference for local styles, with over half of the vessels (54.3 percent) belonging to a characteristic Upper Maeander shape and with an extremely small proportion having a shape characteristic of any other region (table 3). From this, there seems to have been considerable standardization in both vessel style and shape (fig. 6). This suggests an unusual standardization in Level III society of aesthetic preferences. Only a very narrow stylistic range seems to have been considered appropriate for fineware ceramics, which implies that there were wellestablished ideas of how objects meant for display should look and standard visual means by which status could be signaled. Coupled with the continuity in ritual practice already mentioned, this implies stability in the social structure, with strong local traditions in cult and established conventions in the aesthetics of material culture intended for display. There is therefore strong evidence for wealth, stability, and also some level of social hierarchy in Level III Beycesultan, although little can be known about the nature of these social structures. Level III Enactments of Community The remains of Level III Beycesultan do show signs of a coherent sense of community identity at the site, constructed through social practices and enactments of community. Specifically, these enactments of community seem to have taken the form of communal dining activities. Feasting or commensality can be inferred for two reasons. First, fineware dining vessels constitute a conspicuously large proportion of the overall ceramic assemblage recovered (table 2). Even allowing for the biases in the excavation records that skew the assemblage in favor of decorated finewares, the ceramics from this phase show an unusually large amount of shapes associated with drinking, serving, and pouring. Second, these dining vessels were found concentrated in specific areas within the site, implying their use in relatively large-scale communal dining activities. Most notably, almost half of all dining vessels were found within one out of the six structures that were excavated (Building iv), and no dining shapes at all are recorded in some of the structures (table 5). The combination of this notable emphasis on dining shapes in the assemblage with their spatial concentration within the site suggests that some kind of feasting or at least commensality was practiced in LBA Beycesultan. Page 105 → As mentioned in chapter 5, feasting offers a potential venue not only for the expression of group solidarity but also for the competitive articulation of individual identities. In this case, it appears that group solidarity, rather than differentiation, was emphasized in Beycesultan's communal dining activities. This is suggested by the high level of homogeneity Page 106 →within the dining assemblage, which conforms to a strictly standardized set of shapes (Mellaart and Murray 1995), as well as being focused on two very similar decorative styles. It therefore seems that in LBA Beycesultan, commensality brought together groups of people and stressed their similarity by employing standardized and homogeneous material culture. Differences between group members would have been played down in such a context, and the unity between them would have been played up. This evidence for enactments of community therefore suggests that there would have been a conscious sense of collective us at Beycesultan. But what was the social rationale of this community identity, and who was the external them against which the internal us was defined? The representation of external Other through the deliberate use of material culture styles does give us some clues from which to answer these questions.

Level III Representations of the External Other The evidence from the small finds shows that the Level III community actively participated in the long-distance networks of trade and exchange that spanned the Mediterranean and Near East during the LBA. As discussed in chapter 6, this was not a focused, directional trade between point of production/origin and point of consumption /deposition. Rather, a wide variety of objects were transported together in mixed cargoes on circular, “tramping” trade routes, being bought, sold, exchanged, recycled, gifted, and lost along the way as merchant ships and caravans traveled along winding interregional routes. Objects obtained through these networks might therefore not necessarily have borne resonances of their original point of origin but might have been generally significant, instead, of external connections and LBA cosmopolitanism. The beads made from rock crystal, faience, and carnelian are examples of this, as is the Mycenaean stirrup jar (fig. 7). The use of colored beads as eye-catching exotic objects in the socially significant context of personal adornment suggests that they would have signaled these general external links and cosmopolitanism. Similarly, the Mycenaean stirrup jar is unique at Beycesultan and would probably have been recognizable as distinctively foreign, but as stirrup jars were very broadly distributed across the Mediterranean and a recurrent feature of LBA trade networks, it would probably have evoked a pan-eastern Mediterranean koine of semiluxuries more than specifically Aegean connections (Mountjoy 1993; Sherratt 1999; Wijngaarden 1999, 2002). Similarly, the thirty-six bronze items found across the Level III settlement suggest merely external contacts, not the symbolic signaling of external associations. Bronze as a material would have necessarily been imported to the site, either in finished products or as its raw materials, tin and copper. The discovery of a stone mold for making metal tools in Area R implies some on-site production of metal artifacts (Cat. no. B1–51), and this combined with the relative commonness of bronze and its use for practical tools such as chisels and awls as well as prestige objects does not suggest that bronze as a material evoked external relationships, although it may well have been used to signify other social identities, such as status. While the Level III community was clearly involved in the bronze trade, which was an important part and perhaps even a driving force of pan-Mediterranean trade (Sherratt 1994; Stos-Gale 2000), the material of bronze does not, in itself, seem likely to have evoked associations with external contacts. Page 107 → Another imported material present in Level III in not insignificant quantities is white marble, which does not naturally occur in the Çivril valley. The Maeander plain to the south and west, however, is well known for marble. Aphrodisias, in the Dandalus valley, was famous for its quarries during the Roman period (Ponti 1996; Rockwell 1996), and the area around Page 108 →Denizli is the center of a major marble industry today. Although no work has been done to determine the likely source of the Beycesultan marble, it is likely that it primarily came from these Maeander quarries to the immediate south, given their proximity to the site. While marble does seem to have been associated with potentially symbolic objects such as dagger pommels and anthropomorphic figurines, the material does not in itself seem to have primarily indicated external contact, as it was used to make objects in styles that belong characteristically to Beycesultan itself and have a long history at the site. The dagger pommels are of a type common in the MBA levels (Mellaart and Murray 1995, 121–22), and one of the figurines found in the Area R shrines is of a type that has been closely associated with the shrine area since the EBA (Cat. no. B1–41). Marble therefore does not seem to have carried any external connotations in itself. There is little sign that any of these objects would have recalled any specific external associations for the community here. Instead, they would have been reminiscent of general networks of trade and of the cosmopolitan international elite culture of the time. These objects do not, therefore, formulate a coherent image of a specific external Other or shape a representation of the them against which the community us may have been defined. However, two elements of the Level III assemblage do act in this way: one specific marble item and the more general trends in ceramic decoration. The marble item in question is a figurine found in the shrine area (Cat. no. B1–40). While the nature of marble as a material does not imply any specific representation of the Other in itself, the form of the figurine does. Its form is not one that has occurred at Beycesultan before, despite the community's long use of marble anthropomorphic

figurines. Stone figurine types previously in use at the site are limited to two main forms—the violin shape with schematic protruding arms and the figure-eight shape with rounded body and head. This figurine, however, deviates from the Beycesultan tradition and takes the form of a semicircular body with a stalk neck. This form—familiar from several EBA and MBA sites in the wider Upper Maeander region, including Kusura (Lamb 1936, 29), Aphrodisias (Joukowsky 1986, 220), and Elmali-Karata (Mellaart and Murray 1995, 122)—has no history at Beycesultan. Its unusual style would therefore have marked it out, making it visibly “different,” especially in the context of Beycesultan's conservative cultic traditions. The introduction of an object with external associations into the traditional ritual space of Area R would have been a significant social act, as this space was highly socially charged. The presence of a selected type of external Page 109 →Other in such a context suggests a statement of affinity with that Other and a limited acceptance of it. The Other in question, however, is a relatively local one. The figurine's style is connected with the broader Upper Maeander region—the area in which Beycesultan was located. The presence of the figurine within the shrine building therefore indicates a willingness to incorporate Upper Maeander styles, even if they are not focused solely at Beycesultan itself. Thus a sense of broadly local affiliation is also evident in the Level III pottery. The decorative style categorized here as Type C1 is closely (almost exclusively) associated with the local area around the Çivril valley, and the overwhelming dominance of the style in the ceramic assemblage indicates the strength and importance of this association for the Beycesultan community (table 4). In addition, a substantial proportion of the vessels from Level II are made in characteristically local shapes such as chalices and fruitstands. while only 3.7 percent of the assemblage consists of recognizably central Anatolian or Aegean shapes (table 3). In particular, the choice of vessels in both local styles and shapes for use during feasting is especially significant, as feasting in Level III was an arena used to stage enactments of community. Not only were the vessels used in feasts all very standardized, but they were standardized in an emphatically local style that recalled associations exclusively of the local and Upper Maeander area. Another ceramic trend that might be thought to have significance is the popularity of metal skeuomorphism in the fineware shapes. Decorative features such as ribbing on stems, carinations in bowl profiles, grooved or incised decoration, and the addition of clay nodules to a vessel's surface to give the impression of metal rivets are all standard features of many Level III shapes (Mellaart and Murray 1995, 2). The deliberate imitation of metallic features at Beycesultan could be read as related to the status of metal vessels in central Anatolia and the Hittite Empire. It has been remarked that Hittite ceramics are particularly simple and do not seem to have been markers of any type of status at all, leading to the suggestion that high-status vessels in the Hittite Empire were made exclusively from metal (Gurney 1964, 196). If this was the case, metal skeuomorphism at Beycesultan might imply that the community was trying to emulate high-status Hittite practices, which would constitute a deliberate statement of self-association. However, there is very little evidence for Hittite use of metal vessels beyond the ritual sphere (Reeves 2002, 201), and there is no reason to assume that the skeuomorphism at Beycesultan was linked to central Anatolia. I would argue, therefore, that metal skeuomorphismPage 110 → is an aesthetic preference particular to Beycesultan and reinforces the sense of a highly localized geographical affinity. Level III Community Identity and Its Social Rationale This extremely localized sense of community identity should be seen in its wider historical context. During the early years of the thirteenth century and around the time of the construction of Level III, parts of western Anatolia were attacked by the Hittite king Mursili II and, according to the Hittite documents, subsumed into the Hittite sphere of political influence as vassal states (see chapter 6). While official Hittite accounts of Mursili's “Arzawa campaigns” cannot always be taken at face value because of their status as pieces of imperial rhetoric, we can assume that there was some Hittite military intervention in the region around this time and that, following this, several western Anatolian statelets and principalities entered into official alliances with the Hittite Empire. David Hawkins has convincingly argued that the Upper Maeander region was one of these allied regions, Kuwaliya (1998). While there is no certain way of confirming this argument, it is highly likely that Beycesultan, situated in the Upper Maeander region, on a potentially important route between the central Anatolian plateau and the Aegean coast, would have been one of the western Anatolian communities that came under Hittite influence at this time (see fig. 2 in chapter 6).

The adoption of a strongly local community identity at Beycesultan during the period when a political alliance uniting several parts of western Anatolia was being broken up and when its constituent parts were being drawn closer to the Hittite sphere would have been a clear political statement. Although it would have been very likely to have come under strong Hittite influence politically and to have been exposed to Hittite material culture on a practical level, the community at Level III Beycesultan did not choose to articulate these contacts in its material culture. The standardized set of Hittite pottery, which can be convincingly related to patterns of Hittite economic control and production (Postgate 2007), are completely absent. The Hittite language of power—the artistic motifs, visual styles, and cult practices used in the imperial heartland of the Anatolian plateau—was adopted enthusiastically in southeast Anatolia and the Amuq as convenient for expressing elite status (M.-H. Gates 2001; Gunter 2006; Hawkins 1995) but is completely absent from Level III Beycesultan (Glatz 2006, 151–54; Mellaart and Murray 1995, 108; Seeher 2005, 40; Sagona and Zimansky 2009, 285). Page 111 → Similarly absent are any signs of connections with the west. No objects have been found that would have recalled associations specifically with western regions or other parts of what would have been the Arzawan confederacy. It is perhaps more surprising that there is so little sense of solidarity with western Anatolian communities further to the west during a time when Beycesultan is thought to have been engaged in a close political and social relationship with them and then, later during this phase, at a time when this close alliance is thought to have been forcibly partitioned by their shared enemy. The lack of any explicit material reference to the west during both the peak period of the Arzawan confederacy and its breakup might suggest an inherent weakness in the confederacy. It might imply that the alliance was based on calculated political expediency in the face of a large and powerful empire nearby, rather than on an independent sense of regional solidarity. Instead, at Level III Beycesultan, it seems that the population favored symbolic objects that bore specifically local associations, particularly favoring them for use in situations where a sense of community was enacted—in communal dining. This “local identity for local people” can therefore be seen as a reaction to the historical circumstances, emerging at a time of open hostility between western Anatolian groups and their powerful imperial neighbor to the east. It is a reaction to Hittite imperial influence, while it is also a sign that the Arzawan confederacy may not have been as close an alliance as has previously been thought. If Beycesultan is to be identified as lying within Kuwaliya, it seems that the administrative separation of Kuwaliya from other areas may not have been an arbitrary geographical designation imposed by Mursili on an unwilling population. Kuwaliya may well have been a close-knit locality already with a strong sense of cohesion and collective identity. Certainly, the us of the community identity in Level III Beycesultan seems to have been related to the Upper Maeander area exclusively, and very little acknowledgment was made of an external them.

Level II The Remains of Level II The following phase, Level II, dates to the final decades of the LBA and the first few of the IA and shows notable developments from Level III in terms of both internal dynamics and external interactions (fig. 8). Level II Page 112 →shows signs of an increased level of prosperity as a whole and an ever-greater sense of social differentiation and centralization in the control of resources. Again, there is continuity in terms of ritual activity in Area R (Lloyd 1972, 27–32) and the overall urban layout of the east mound (Lloyd 1972, 9–17), as well as an increase in architectural elaboration in both of these areas. In Area R, the enlarged shrine building features more complex entrances and anterooms with additional architectural installations. On the east mound, there is more regularity and an overall higher standard in construction methods, and ornate domestic complexes appear alongside more standardized domestic units. Some have central “public” rooms, like the megaron Buildings vii and viii; others have their own courtyards and lustral areas, such as the house south of Building vii (“Megaron B”) and Building ix (dubbed by the excavators the “Little Palace”). On the west mound, domestic units are of a more standardized nature but are still multiroom complexes with their own courtyards, and one has what appears to be a household shrine (Lloyd 1972, 19–21).

This greater sense of general prosperity is accompanied by evidence for more social inequality and hierarchy. The domestic units, for example, show substantial variation in terms of both architectural elaboration and size, with the largest house occupying more than eight times the total floor space of the smallest (see table 6). This inequality suggests more marked social differentiation than in Level III, enshrined in the built environment. Centralization of resources is also suggested by the substantial storage facilities to the east of Building ix. The excavators interpret these two rooms as shops—one for wine, the other for food (Lloyd 1972, 12–13). The rooms, however, are just as likely to be centralized stores controlled by Building ix. Whether the excavators' guess is correct or not, the centralized storage of foodstuffs on a scale far beyond the domestic suggests a high degree of social complexity. Yet another sign of hierarchy is the stabling complex for horses in a prominent central location on the east mound (Lloyd 1972, 15–17). Costly to procure and to keep, horses were a sign of extremely high status in the eastern Mediterranean during the Bronze and Iron Ages (Dietz 1999; Hyland 2003) and were especially highly prized in the LBA, when chariots and chariot warfare were first introduced (Moorey 1986). Together with the horses they would have housed, the stables, which occupied the eastern half of the insula block between the two streets, are therefore a sign of wealthy elites within the Level II population. Status distinctions are implied once again by the presence of sealings and seals, both stamp and cylinder (Cat. nos. B2–75, B2–53, B2–74), of which the stamp seal is particularly ornate. Small finds likely to be considered prestige items are also found again in Level II, including necklaces made from shells (Cat. nos. B2–54, B2–55) and frit (Cat. no. B2–56), metal jewelry (Cat. nos. B2–31, B2–32), serpentine axheads (Cat. nos. B2–38, B2–39), and ornamented daggers (Cat. nos. B2–33 to B2–37). In terms of ceramics, there are very few changes between Levels III and II, with the range of vessel shapes in use remaining largely the same (Mellaart and Murray 1995, 22), although there are some changes in the proportions of Page 114 →different shapes within the overall assemblage (table 2). However, there is a notable increase in the use of metallic slips, especially for drinking shapes (Mellaart and Murray 1995, 21–22), heightening the already metallic nature of the pottery assemblage (table 4). Page 113 → Level II ended in a burnt destruction, which accounts for the larger number of finds from this level than any other (all other levels seem to have ended peacefully, with abandonment and “cleaning”; Lloyd 1972, 5). Indeed, the destruction of Level II seems to have been violent, as projectiles and items of weaponry have been found across the site, and eight unburied human skeletons were discovered in the “shop” area, which the excavators suggest were the heaped up bodies of those “clubbed to death” (Lloyd 1972, 12). The conflagration seems to have spread over much of the settlement, including the domestic areas on both mounds and the shrine in Area R. While there is a temptation to compare the burnt destruction with those elsewhere in the eastern Mediterranean at the end of the LBA, the revised chronology outlined earlier in this chapter suggests that the destruction of Level II probably did not occur until the middle to late twelfth century, some decades after the fall of the Hittite Empire and the collapse of the Mycenaean palace states. It seems, therefore, that the rigidly hierarchical, highly organized, and visibly wealthy society of Level II Beycesultan survived the collapse of nearby states, continuing on into the IA with its social and economic structures comfortably intact and only suffering a burnt destruction later in the IA itself. Level II Enactments of Community There is relatively little evidence for deliberate expressions of group solidarity in the hierarchical society of Level II. Vertical social differentiation and distinctions of rank seem far more important than community togetherness. Although dining vessels still constitute a substantial proportion of the overall ceramic assemblage (table 2), this is not beyond the proportion that might be expected given the bias in recording practices admitted by the excavators, and it is notably less than the proportion of dining vessels in the previous level, Level III. Similarly, while some of these vessels have been found in clusters that suggest commensality, these clusters seem to have been relatively evenly spread between several structures (table 5). While communal dining may have been going on at several different locations, there does not seem to have been any single focal point for this, unlike in Level III. Unlike in the previous phase of occupation, Page 115 →there is no indication of large-scale group activities that encouraged a sense of broad-based corporate solidarity. Alongside this lack of evidence for practices of group affiliation, the ceramic assemblage is less homogeneous and

standardized (table 4). The repertoire of vessel shapes remains more or less the same, but the decorative schemes on finewares diversified. As well as the warm-colored schemes that so dominated the previous phases (Types C1 and C2), lustrous schemes (Types A1 and A2) now also become popular. Although this does not represent a particularly wide range of different styles in use, it is nonetheless significant that elite ceramics were no longer visually homogeneous. While Level III vessels encouraged a sense of similarity and standardization, this leveling factor is completely absent from the Level II assemblage. In Level II Beycesultan, therefore, there is no longer any obvious evidence for a sense of coherent community identity being enacted. Instead, there seems to have been a much more structured social hierarchy, dividing households along a rising scale of status and wealth. Overall, it seems that people in this phase were much more concerned with differentiating themselves from each other than with emphasizing their commonalities. Level II Representations of the External Other Beycesultan continues to have a wide range of external contacts in Level II. The community was still involved in pan-Mediterranean trade networks, which allowed for the continued import of raw material such as metal ores for local metals industry, as well as exotic luxury items such as frit beads (Cat. no. B2–56) and substantial numbers of cowrie shells (Cat. nos. B2–54, B2–55, necklaces). At the same time, Beycesultan seems to have maintained close contacts with areas to the south and west within the Maeander region, as attested by the sustained import of marble. Beycesultan also had regular connections with central Anatolia at this time, as evident especially in the appearance, toward the end of the level, of the rope-impressed coarse wares (already mentioned in the context of dating this phase). As in the previous level, therefore, several of the finds do imply external contacts but are nonetheless not likely to have been part of deliberate representations of the Other. Several objects made from distinctive types of stone are among these items. These include several items made from serpentine (Cat. nos. B2–38 to B2–41, B2–47), the spout of a stone vessel made from delicate blue limestone Page 116 →(Cat. no. B2–52), and a perforated flint object, perhaps a whetstone (Cat. no. B2–46). The stones from which these objects were made are relatively rare and may therefore have been associated with their place of origin. However, the sources of these stones cannot at present be traced, and there is no real way of knowing whether the source would have been known in Beycesultan or recognized immediately from the material. Another group of objects that offer clear evidence for external links but are not directly relevant to this consideration of community identity are the weapons found in the Level II destruction layer. These weapons were not used by the people of Beycesultan as a means of representing and negotiating their external relationships but were instead left in situ following the violent destruction of the settlement at the end of Level II. The weapons do not point clearly to any single area of origin—the excavators have drawn parallels in the style of the arrowheads with those found at places as diverse as Mersin on the south coast, Alacahöyük in central Anatolia, and Hospital Tomb No. 3 at Knossos (Lloyd and Mellaart 1955, 91). However, it is also noted that the leaf-shaped arrowheads found at Beycesultan are notably different from the barbed types used by the Hittite armies, meaning that the destruction of Beycesultan was not due to Hittite aggression. The identity of Beycesultan's attackers, therefore, remains a mystery. The ceramic assemblage, in contrast, does seem to have been used for representing the external Other. Although the repertoire of shapes in Level II shows an emphasis on local forms that is similar to that in Level III (table 3), there is a marked change in decorative preferences. A dramatic decline in the proportion of Type C1, warmcolored burnished wares, is matched by a corresponding increase in the popularity of Types A1 and A2, lustrous wares (table 4). The proportion comprising lustrous wares does rise from 15.7 percent in Level III to 32.5 percent in Level II, with both types roughly doubling in terms of the percentage of the overall assemblage they account for (Type A1 going from 3.1 to 7.7 percent, Type A2 going from 12.6 to 24.8 percent). The increase in the popularity of Type A1 suggests a conscious recognition of links with the northwest around the Hermos river valley, as the general aesthetic of gold- and silver-washed pottery was probably associated with this area (see chapter 6). The style first appears at Beycesultan in Level III, but its notable increase in popularity in

Level II is significant. Type A2, constituting the single largest group in the Level II assemblage, is also interesting in this light. Although clearly influenced by the micaceousPage 117 → washes of Type A1, these coppery or red lustrous wares do not appear in the northwest at all but are found almost exclusively in the immediately local area around the Çivril valley. As a visual style, Type A2 combines the lustrous finish of Type A1 Hermos valley wares and the warm colors of Type C1, the preferred decorative type of Level III significant of a traditional Çivril valley aesthetic. The popularity of Type A2, therefore, may have stemmed from the fact that it referenced both the Hermos silver and gold lustrous style and the local tradition. Type A1 itself would have fitted in comfortably with the existing ceramic preference for metallic imitation at Beycesultan, an external influence that could have been easily integrated into existing internal tradition. Therefore, while the increase in lustrous wares does suggest a reorientation toward areas to the west, it appears that this was a selective reorientation, not compromising existing ceramic traditions. This affinity with areas to the west of Beycesultan may also be indicated by a part of a boar's tusk helmet found in Trench J (Cat. no. B2–91). Such helmets are synonymous with Homeric warfare and epic poetry but are also a routine part of Mycenaean armory. Examples have been found in a number of Mycenaean tombs, as well as in Aegean figural art (Dickinson 1994, 201–7). An association between boar's tusk helmets and the Mycenaean Aegean in general is suggested by the apparent existence of an artistic convention for representing Aegean warriors in standardized dress including greaves, kilt, and boar's tusk helmet (Bennet and Davies 1999; Neimeier 1999, 150–51; Schofield and Parkinson 1994). The discovery of part of such a helmet at Level II Beycesultan might therefore suggest a deliberate recognition of connections with the west as far as the Aegean coast. While the fragment must have signified connections with the west, it is uncertain what types of connection these might have been. Such an item might have been kept as a statement of external association but may equally have been a battle trophy, articulating a social relationship of hostility rather than affinity. In this case, it is impossible to ascertain which. Another object that is likely to have had strong external associations is the stone stamp seal (fig. 9; Cat. no. B2–53). The stamp motif is of a crouched griffin, surrounded by two concentric circles in what is known as “ladder design.” Both the motif and the form of the seal have been compared by Murray to MBA central Anatolian examples (Mellaart and Murray 1995, 123). While it is probable that the style would have been recognized as linked to central Anatolia, it may not, however, have been associated with the Hittite state, as the central Anatolian comparisons all date to before the Hittite imperial period. Nonetheless, a seal is, by its nature, an object with potential to be used symbolically, as it would have been used for official and administrative purposes and is suggestive of both political structures and personal status. The griffin seal, with its strong visual links to central Anatolia but not to standard glyptic styles of the Hittite state, is interesting given its context of the final phases of Level II. The seal would have been in use in the decades following the collapse of the Hittite state. It is likely to have had a very particular social meaning, used at this moment in time. This is discussed in more detail in the next section of this chapter. Page 118 → The seal is also noteworthy because it runs counter to the prevailing trend at Beycesultan that looks westward rather than eastward. It seems, therefore, that the people at Beycesultan chose to use objects that were reminiscent of a range of different external Others. This is perhaps unsurprising. There is no clear evidence from this level for the inhabitants of the site ascribing to a single, coherent sense of community identity during this phase. Different households and individuals instead strove for distinction from each other and emphasized hierarchical differences of wealth and status. In such circumstances, the evoking of external contacts and the choice of which contacts to evoke would also have been means of negotiating internal differentiation. Different groups within Beycesultan Page 119 →chose to use objects that recalled different external associations, as part of their efforts to highlight their difference rather than their similarity. Level II in Its Historical Context Level II spans from the end of the thirteenth century through the early to middle twelfth century, an unstable time

in the eastern Mediterranean and particularly in Anatolia, around the period of the LBA-IA transition. During these unsettled years, which saw the destabilization and collapse of several eastern Mediterranean states and the inevitably confused aftermath, the internal dynamics at Beycesultan are informative. The sense of conscious community identity in Level III gave way in Level II to an emphasis on internal status differentiation. In addition, the exclusively local focus of Level III gave way to the selective integration of external Others in Level II. The removal of the external political threats can largely explain both of these shifts. During the time of Level II occupation, Beycesultan would no longer have been in the shadow of its threatening imperial neighbor, the Hittite state. External Others were no longer an imminent danger, and relations with other western Anatolian sites may also have been less fraught and pressured as a result. This changed historical context would have removed some of the urgency for community solidarity and would have allowed the inhabitants of Beycesultan to direct their differentiating tendencies inward rather than outward. As it was now less important for everyone to pull together in the face of an external threat, people became more concerned with articulating differences in rank and status. Overall, people at Beycesultan became concerned with distinguishing themselves less from the Other as from each other. The shift from an aesthetic range that focused narrowly on the local can also be related to the changing historical circumstances. As external pressure was reduced, external styles and resonances became less threatening, and the inhabitants of Beycesultan were freer to incorporate a number of external styles into their repertoire. Most notable here are the western associations that would have been recalled by the decorative schemes of fineware ceramics. The idea of an area previously under Hittite influence now looking westward fits in well with current interpretations of the end of the empire. These suggest that the power of the central government in Hatti declined as that of neighboring states that had once been allied to it increased. For example, both Tarhuntassa in southern Page 120 →Anatolia and Karkemish in northern Syria, once client states of the Hittite Great King, became increasingly independent in the final years of the thirteenth century, with the rulers of both styling themselves as “Great Kings” in their own right (Klengel 2002, 108; Mora 2003). David Hawkins has argued that this phenomenon also occurred in western Anatolia, where he sees a political resurgence in the old Arzawan principality of Mira (2002, 150). If his location of the territory of Mira between the Cayster and Hermos rivers is correct, the growth in popularity at Beycesultan of gold and silver lustrous wares—a style closely associated with this area—would imply that the people of Level II Beycesultan were signaling some kind of affinity with the growing political force of Mira (see fig. 2 in chapter 6). The inhabitants of Level II Beycesultan may therefore have been aligning themselves with a growing political power in the west in the face of a weakening Hittite state. But they did not adopt western influences uncritically. They appropriated only those ceramic styles that would sit well with existing local traditions, opting for metallic lustrous slips that would emphasize the already-metallic character of their pottery. In addition, they adapted the western practice of the lustrous slip by combining it with previously popular red burnished wares to create the coppery lustrous ware that is rarely found outside the Çivril valley area. Alongside this, they also continued to use their own local decorative traditions. So while associations with the west were being expressed at Level II Beycesultan, this was not done at the expense of the local identity so important in the previous period. However, the appearance of the griffin seal suggests that the Hermos area was not the only external area from which people at Beycesultan chose to borrow. Appropriation of a central Anatolian style would have been somewhat conspicuous given the overall westward-looking trend of the time. It is possible, however, that the collapse of the Hittite Empire had rendered more general central Anatolian styles “safe” for appropriation, perhaps no longer loaded with hostile political connotations. Nonetheless, that a central Anatolian style was thus embraced is interesting. In addition, the choice of a seal itself is particularly interesting, as seals are socially significant objects with connotations of status and administrations. In conclusion, the griffin seal is a reminder that social actors at Beycesultan were seeking, above all, to mark themselves out from the crowd. This was not a coherent and unified community. It was a selection of individuals seeking to differentiate themselves from each other. Page 121 →This seal certainly would have been a potent, if potentially controversial, marker of difference.

Level Ib

The Remains of Level Ib After the destruction of Level II in the mid-twelfth century, the community appears to have recovered quickly, with the Level 1b settlement being built largely on the same street plan as its predecessor (fig. 10; Lloyd 1972, 17–18). In addition, the economy continued to support long-distance trade and imports of several luxury items, such as a faience figurine (Cat. no. B3–58), an ivory stamp seal (Cat. no. B3–59), marine shells (Cat. no. B3–57), metal vessels (Cat. no. B3–05), and a serpentine ax (Cat. no. B3–15). However, the structured social asymmetry in the Level II community seems to have disappeared, to be replaced by a sense of flexibility or instability in the social order. The domestic units of Level Ib are much more equal both in their level of architectural elaboration and in the amount of space that they occupy. Although a small sample, the four preserved houses are all relatively small structures (two to three rooms) with a single main room for most household activities, and all cover a similar amount of total floor space (table 6). There are no clear markers of any of these households being of higher or lower status than any other. Significantly, in Level Ib, there are no obvious public or central structures like the shrine buildings or the stable complex of the previous levels, suggesting less centralization and less-structured social organization. There was, however, a large open area to the east of the houses, which was plastered all over with a thick layer (five to ten centimeters) of white gypsum and appears to have been used for ritual purposes. Two small circular pits had been sunk into the plaster at what would have been the exact summit of the mound, filled with fragmented animal bone, and covered with rough slabs of stone (fig. 11). There was, then, some central focus for the community at Level Ib Beycesultan, although activities associated with this kind of space necessarily imply a different set of social relations than those of the shrine building of Levels III and II. A large, open space such as this, involving no built structures, would encourage inclusivity, public display, and the active performance of status. This contrasts starkly with the narrow shrines with anterooms of earlier periods, which imply exclusivity and status through control of selective participation. While the twin shrine buildings suggested symbolic control by an established elite, the open plastered area of Level Ib indicates the active construction and public contestation of unstable statuses. Page 122 → The signs of decentralization and destabilization in the architectural remains are matched by signs of status competition in the small finds. A new emphasis on jewelry and bodily adornment suggests an increased focus on the body and personal status of the individual. Of the sixty-two objects known from Level Ib, ten—six metal rings, a shell necklace, a bone pendant, and two beads—can be classed as jewelry. The ten objects account for 16.1 percent of all objects registered, and all of them can be linked to personal bodily adornment (table 1). This is a much larger proportion of the small finds assemblage than in the highly stratified society of Level II, with its rigid structures of social differentiation. There, only four objects, comprising 4.4 percent of the small finds assemblage, could be linked with individual adornment. It should be noted that while eight jewelry objects altogether were found in Level II, four of them (Cat. nos. B2–44, B2–45, B2–46, B2–79) came from the shrine buildings and so should be considered as votive deposits rather than as items for personal adornment. Although these percentages do not accurately represent the complete archaeological assemblage, the selection criteria for documentation would have actively selected for items such as these, and these figures thus offer us a sense of the relative investment in personal adornment in the different levels. The increased interest in the body of the individual between Levels II and Ib suggests a concern with the active creation and contestation of status within a more flexible and unstable social structure. Page 123 → Status competition is also indicated in the ceramic record, as there is Page 124 →a sudden diversification of the decorative types in use. All decorative styles included in the typology are present, save for Mycenaean, Lydian, Phrygian, and East Greek styles, none of which are contemporary with the dates proposed for Level Ib (table 4). Whereas there were distinct aesthetic preferences in fineware style in previous levels, Level Ib shows a sudden

proliferation of different types of decoration, bearing resonances from many different places. As well as the previously popular lustrous wares (Types A1 and A2) and warm-colored burnished wares (Type C1), there is a substantial increase in the amount of linear painted pottery (Types B1 and B2) and black and grey wares (Type D), as well as the introduction, for the first time, of pale slipped and white wares (Type E). In addition, Mellaart suggests that the burnished wares themselves diversify in terms of color, including what Mellaart evocatively describes as “chocolate brown,” “deep plum red,” and “rich camel-leather buff,” although this is not readily verifiable from the records and remaining stored pottery (Mellaart and Murray 1995, 56). Overall, there seems to be no dominant single fineware style, and the wide variety of aesthetic schemes in use suggests competitive display and innovation. The inhabitants of Level Ib were therefore still prosperous and still maintained a wide range of external contacts across the eastern Mediterranean. Level Ib was not, therefore, a period of “Dark Age” decline and backwardness, as assumed by the excavators (Lloyd 1972, 5). The main differences between Level Ib and the preceding phase was not in its level of attainment along a putative scale of civilization but, rather, in the way society was structured. Compared with earlier periods, the inhabitants of Level Ib would have been much more socially mobile; as social structures were flexible and open to competition, status could be both acquired and lost. Level Ib Enactments of Community Competition and fluidity therefore characterized Level Ib, in contrast to the structured hierarchy of the previous phase. This lack of rigid social distinctions and hierarchy does not necessarily imply a sense of collective group identity or corporate solidarity. The evidence seems to suggest that social distinctions were just as important in Level Ib as in Level II Beycesultan—the difference is that status was more flexible and open to negotiation. But while the inhabitants of Level Ib Beycesultan were certainly very concerned with differentiating themselves from each other, the remains of this level suggest that there were some potential arenas for the construction of a group identity as well. Page 125 → The first of these is the open ritual area previously mentioned. This is a large open space that could have accommodated a substantial number of people without any obvious signs of spatial separation or distinction. This area, therefore, could potentially have been used for staging enactments of community. Similarly, there is also evidence for large-scale feasting activities during this phase, and these could also have offered opportunities for enactments of community. As in Level III, the fineware dining vessels of this phase are concentrated into two main deposits, with more than half the dining vessels found together in a single area (table 5). This implies that while communal dining during the Level II occupation had only been carried out on a small scale, the inhabitants of Level Ib feasted in much larger groups. Feasting and ritual activities could potentially therefore have both been arenas for the enactment of community in Level Ib Beycesultan. However, given what we already know about Level Ib, it seems unlikely that feasting and ritual served to forge a sense of communal solidarity. Instead, they seem more likely to have been the settings for social competition, public display, and the active contestation of status. The evidence from the small finds and the diversity of the ceramic assemblage suggest that distinction and innovation were more of a concern than similarity and cohesion. Therefore, while there were two potential venues where community identity could have been constructed, it seems much more likely that they were used to express division rather than unity. Level Ib Representations of the External Other Competition and plurality also seem to have characterized the way in which the external Other was represented in Level Ib Beycesultan. Notably, the material record shows how a wide range of external influences were embraced, signaling recognition of a wide range of different external Others. These different Others are likely to have been pitted against each other as part of the competitive ethos of the time, and visual cues recalling different external Others may have been used as an additional means of articulating social differentiation.

The best illustration of this is the sudden glut of diverse styles in fineware decoration. The variety of decorative types and the multiplicity of different color schemes now in use have already been mentioned, in the previous section. It is also significant, however, that these different types would have recalled different external associations. Previously popular styles continued in use during this phase, with the local Types A2 and C1 Page 126 →still accounting for a substantial proportion of the overall assemblage. As well as this, the inhabitants of Level Ib also continued to favor the northwestern lustrous style that was so notable in Level II, and the proportion of Type A1 vessels in the assemblage increased yet further. In addition to this, however, connections with other parts of the Upper Maeander area directly to the west were articulated through the use of Type B1 pottery. Type B1, linear painted decoration on a lustrous surface, is strongly associated with Aphrodisias and would likely have recalled this area for the inhabitants of Beycesultan. There was also an unprecedented rise in the proportion of Type D, grey and black wares. This would have evoked a sense of the far northwest, particularly the Troad. In addition, the appearance of Type E pottery with pale and white surfaces would have evoked the southeast, Cilicia and beyond. Overall, by diversifying in form and decoration, the ceramics of Level Ib would have recalled a diversified range of external Others for those who used them, taking in the Troad to the far northwest and Cilicia in the southeast as well as the more familiar areas of the Hermos valley and the Upper Maeander styles of both the Dandalus valley and the immediate Çivril valley locale. The only nearby area that leaves no trace on the fineware assemblage is central Anatolia. This is highly significant and may suggest that even well after the fall of the Hittite state, there was still some social distancing between western Anatolian communities and those of the plateau. One object that could be presented as an argument against this is the fragment of a libation arm found in the open plastered area of Trench M (Cat. no. B3–119). The appearance and popularity of libation arms in Anatolia were closely linked to the expansion of the Hittite Empire into northern Syria in the mid-fourteenth century (Eriksson 1993, 129), and they are found in large numbers in the thirteenth-century levels at Bogazköy (Fischer 1963, plate 122:1102, 1124; plate 124). Their LBA distribution, however, is very wide, covering much of central and southeastern Anatolia (Eriksson 1993, 129–34), and their continued use in the IA beyond central Anatolia is attested by finds from Kilise Tepe (Postgate and Thomas 2008, 121; Postgate, personal communication). The discovery of a libation arm in the open plastered area suggests a ritual context and therefore a highly socially charged one, especially since the shrine buildings in Area R had gone out of use. We cannot be sure, however, whether the libation arm would necessarily have carried central Anatolian connotations for the Level Ib community. Instead, this object might be better interpreted as a sign of participation in what might have become a general Anatolian cultic tradition, rather than one linked specifically to any single region. Page 127 → The fragment of an imported faience figurine is equally unhelpful (Cat. no. B3–58), even though it recognizably belongs to a Near Eastern style, with Murray suggesting a close parallel from Ebla in Syria (Mellaart and Murray 1995, 126). Given that Murray's parallel dates to the Old Syrian period, a good four to five hundred years before Beycesultan Level Ib, this import is unlikely to have been viewed as representative of the region in contemporary times. As already mentioned, small but valuable objects were standard cargo on the circuitous, “tramping” trade routes of the LBA and IA, and an item such as this might be considered even more valuable because of its “antique” status in such a context. It is uncertain, therefore, whether the inhabitants of Level Ib Beycesultan would have recognized this style as linked to a specific geographic location such as Syria or as significant of a more generalized idea of maritime connections and the “Near East.” Similarly, the ivory stamp seal from Trench M can also be taken as indicative of no more than involvement with long-distance trade and the generic, rather than the specific, external (Cat. no. B3–59). Neither ivory as a material nor the simple design evokes any region in particular, and the seal is another example of the kind of valuable and generically foreign exotica carried in panMediterranean trade. This supports the general impression of the Level Ib community as one in which the vocabulary of status expression was no longer fixed and standardized but where the definition of elite styles was very much “up for grabs.” The switch to a more public ritual space, the new interest in bodily adornment, and the relative equality of domestic accommodation all suggest a high degree of flexibility and mobility in Level Ib society. In this context,

it is unsurprising that a wide range of external Others were represented, with different competing elements within the community choosing to express rival external associations. The griffin seal found from the final phase of Level II suggests that the seeds of these dissenting external orientations were already laid before Level Ib (see previous discussion in this chapter). It seems likely that the destabilization of established hierarchies across the eastern Mediterranean at the end of the LBA allowed for a new flexibility in intercommunity relationships, and at Beycesultan at least, external affinities were used as symbolic capital in the internal struggle to acquire, confirm, and define status within the community. Level Ib in Historical Context As mentioned earlier, the excavators initially assumed that Level Ib was a period of IA “decline” (Lloyd 1972, 5). However, this phase is better understoodPage 128 → as a time of transformation in the community's social structures, when old social structures were being actively contested, reshaped, and remodeled. While the chronology of Level II suggests that Beycesultan was not immediately adversely affected by the collapses of ca. 1200 BC, the deep changes in social structure and the concept of elite identity that are evident in Level Ib might be seen as some of the delayed or longerterm consequences of the LBA-IA transition. The failure of centralized, rigidly hierarchical, and long-established social systems in other parts of the Mediterranean might have encouraged a more gradual destabilization of long-established social structures and traditional elite identity at Beycesultan, offering an opportunity for status to be redefined and reformed in new ways. In this environment of frenzied status competition, group solidarity and community identity do not seem to have been important at all. The situation in Level II, in which the inhabitants of Beycesultan no longer needed a sense of community solidarity in the face of an external Other, seems only to have intensified in Level Ib. Internal distinctions such as individual rank and identity were a much more pressing concern than any idea of collective togetherness. The social patterns within Beycesultan mirror something of the historical patterns without. The IA in Anatolia saw the rise of a new political order in which no single power dominated. Two significant political powers were already emerging in the Level II period—Mira in the west and Tarhuntassa in the southeast—and these are likely to have continued to be important during much of Level Ib. However, the Cypriot states were also on the rise at this time, coming to dominate trade routes and metal production in many areas of the Mediterranean (see chapter 6). In addition, the Aegean coast of Anatolia saw intense political activity, with the emergence of a Hellenic consciousness in the central and southern coastal areas and with increasing levels of contact with continental Europe in the Troad (see chapter 6). Level Ib Beycesultan was therefore contemporary with a dynamic period of Anatolian history, when political power was dispersed and multipolar and when economic patterns were shifting and geographically diffuse. Flexibility, competition, and diversity characterized both the internal society of Beycesultan and the external historical context in Anatolia more generally. Aside from this diversity, the most marked characteristic of the Level Ib assemblage is the continued absence of any resonances with the central Anatolian plateau. Rather than this being a “hangover” of bad feeling generatedPage 129 → by the Hittite Empire, it seems instead that connections with central Anatolia were simply not as valuable as symbolic capital as were connections with other areas. While there was political activity to the southeast of Beycesultan, central Anatolia does not seem to have played a dominant economic or political role in the wider region at the start of the IA (Hawkins 2002). Visual cues evoking the area around Tarhuntassa in the southeast, such as Type E finewares with pale surfaces, or the area around Mira in the northwest, such as Type A1 silver and gold lustrous wares, may have simply been more potent social symbols at the time. The material from Beycesultan tells us, therefore, not just who the big players were on the IA Anatolian political scene but who lacked political clout as well.

Level Ia The Remains of Level Ia

In Level Ia, what has been interpreted as a single farmstead complex was built over the peacefully abandoned ruins of Level Ib (Lloyd 1972, 18–19). The Level Ia remains consist of a single large megaron complex on the east mound, with the central megaron building flanked by a number of outbuildings set around a courtyard to the west, with a large circular baking oven to the east (fig. 12). The excavators have suggested that the population at this time had shrunk to very small proportions, perhaps only consisting of one extended family (Lloyd 1972, 6). This interpretation seems based on an analogy between this megaron complex and the domestic houses built around a central megaron plan in Level II. However, the Level Ia megaron building is especially large and well constructed, and was built in a more monumental style than the megaron houses of Level II. The megaron itself is free-standing and visually impressive. The first room has an open frontage with pillars, the doorway between the rooms is large (2.4 meters wide) and laid with a timber threshold, and there is a large circular hearth in the middle of the second room. This all suggests that the megaron house was designed with some element of public display in mind, perhaps as a central building for a modest but socially complex community. In addition, the sheer size of the baking oven (ca. three meters in diameter) suggests a scale of food production well beyond the strictly domestic. If these interpretations of the megaron building and baking oven are in-deed correct, this would indicate some sort of return to centralized social structures, albeit perhaps on a smaller scale, in contrast to the social fluidity and mobility of Level Ib. Page 130 → There are relatively few finds securely linked to this disturbed level, but what has been preserved and recorded does provide some information as to the nature of society at Level Ia Beycesultan. For the first time, the finds do not include any exotic imported items such as ivory, faience, silver, and precious stone objects, perhaps suggesting a decrease in the amount or frequency of external trade. Some trade, however, seems to have continued, as attested by the presence of some metal items. These include objects associated with personal prestige, such as two finger rings (Cat. nos. B4–5, B4–7) and an ornate bronze macehead (Cat. no. B4–6). This level was severely disturbed by later building and quarrying activity on the east mound, and in other areas of the site where there may have once been further domestic remains, this level has been completely Page 131 →destroyed. Therefore, we will probably never have a full picture of what was going on in Level Ia. Nonetheless, the little that is known about this phase suggests that Beycesultan at this time may have been home to a complex society, with reasonably stable social structures and reasonable access to resources and manpower. While the inhabitants of Level Ia Beycesultan may not have been quite as wealthy or as well-connected as their predecessors, the quality of their architecture at least suggests that they were far from being backward. Level Ia Enactments of Community The central megaron building and the large baking oven do not only indicate social complexity and stable social organization, however. They also suggest a sense of concern for collectivity and group solidarity that had been absent during both Levels II and Ib. Notably, the megaron is deliberately designed to maximize inclusivity and access. The first chamber is not accessed through a doorway—instead, one whole wall takes the form of an open pillared porch. The doorway from the first into the second chamber is also wide—it occupies two-thirds of the overall wall space. While being impressive and almost monumental, therefore, the design of the megaron also betrays a very conscious concern with inclusivity and open participation. Such a structure actively discourages differentiation, exclusivity, and control of access. This implies that its architects were more concerned with cohesion rather than distinction. Activities undertaken within this space may have had the effect of enacting community identity. Such activities are also implied by the range of ceramics found in the megaron house. The vessels in the megaron represent only a very narrow range of shapes, implying that the building had very specialized uses. Cooking and storage vessels constitute an unusually low proportion of the assemblage (table 2). In contrast, there are at least three different vessels that would have had ritual uses. It seems, therefore, that ritual activities took place in the megaron house and that these included the pouring of libations through libation arms and using fineware dippers. A fragment of a clay altar was also found inside the megaron, lending weight to the interpretation of the megaron

as a space used for inclusive public ritual. The megaron building also seems likely to have been the location of feasting, as well as ritual, activities. A substantial number of the ceramics found are dining vessels (table 2), and these imply commensality both by their absolute numbers and by the proportion of the assemblage that they Page 132 →constitute. Commensality may also be suggested by the large baking oven, as this would have been capable of producing a large quantity of food at once. As remarked on in chapter 5, feasting and ritual activities in themselves do not necessarily constitute enactments of community. They are social activities that can work either to foster a sense of collectivity or to promote a sense of differentiation, depending on the manner in which they are carried out. There is evidence, however, that the feasting and ritual acts undertaken in the megaron building of Level Ia would have been more of the former. First, the nature of the space itself lends itself to enactments of community, encouraging unity, cohesion, inclusivity, and open participation. Second, the ceramic vessels used in the feasting were very standardized in their type and decoration (table 4). It seems, therefore, that the inhabitants of Level Ia Beycesultan did engage in enactments of community set around the central community megaron building and that these included both ritual activities and feasting. Level Ia Representations of the Other Far fewer material traces that evoke other communities and other regions have been found for this phase than in any other, suggesting something of a narrowing in the community's perspective. The pottery returns almost exclusively to local styles, with the warm-colored slips of Type C2 dominating the fineware assemblage and with the characteristic Çivril valley burnished Type C1 also very popular ( table 4). In addition to this concerted focus on local styles, there is a substantial increase in the proportion of Type E decoration in the assemblage, suggesting that while the Level Ia community generally associated itself with the Upper Maeander local area, connections were also being signaled with the southeast. The discovery of a libation arm fragment in the megaron house (Cat. no. B4.44) also suggests some engagement in socially significant practices with external associations. As discussed earlier, however, it is unclear what specific area, if any, the libation arm would have recalled. It is possible, since the only other site known to have libation arms in the IA contexts is Kilise Tepe, that it would have been evocative of connections with the southeast (see previous discussion in this chapter). The other item that could potentially have external associations is the finely crafted brass macehead, also from the megaron house (Cat. no. B4–6). The quality of its workmanship led the excavators to think that it was “ceremonial,” and Page 133 →such an object would certainly have been at least partly intended for display (Lloyd 1955b). This largely local focus should be understood in its eleventh-century context. In the middle of what was once thought of as the Dark Ages (see chapter 6), external interactions and movement were generally less in both scale and volume than either in the LBA or the early phases of the IA. While Beycesultan still seems to have had access in this period to longdistance trade routes that allowed for the continued import of metals, the social significance of external relations and the importance attached to signaling external links may have declined. This interpretation is lent some weight by analysis of vessel shape (table 3). While a reasonable proportion of the ceramic assemblage continued to be made in typically Upper Maeander shapes, the overwhelming majority of vessels are of a generic shape that cannot be connected with any particular geographic location. Instead of preferring shapes that had specific connotations, therefore, it seems as if the inhabitants of Beycesultan opted for shapes that were reminiscent of nowhere in particular. It is possible that the evocative power and allure of the foreign may have waned, as the social structure at Beycesultan may have stabilized (as suggested by the megaron as a central semipublic building) and coalesced around its strong local identity. However, it seems that while the Level Ia community expressed a primarily local identity, it also partly looked to the southeast, as indicated by the increase in the proportion of vessels with Type E decoration and, perhaps, the presence of a libation arm in the megaron. This choice to articulate a relationship with areas to the southeast may be partly due to the economic power of this area in the IA. As discussed in chapter 6, Cyprus grew to be a major player in the interregional trade of the IA and was particularly active in the exchange of metals (Sherratt 1994;

Snodgrass 1994). Metals are one of the few identifiable imports at Level Ia Beycesultan, and perhaps the community's recognition of southeastern styles should be seen in this economic context. Level Ia Community Identity and Its Social Rationale The sense of group solidarity that emerges at Beycesultan in the later IA is difficult to interpret because of the relative lack of evidence for this period. This is true both in the disturbed archaeological record of Beycesultan and also in the history and archaeology of the eastern Mediterranean more Page 134 →generally. This is the period traditionally characterized as the Dark Ages, when social complexity and interregional exchange were thought to have decreased and when and local communities were said to have slid into backwardness and insularity. The remains of Level Ia Beycesultan belie the first of these characterizations. Social complexity is implied by the megaron—such a building could not have been constructed without the existence of some form of social organization (involving authority and specialization, if not necessarily hierarchy). Similarly, it appears that Beycesultan continued to have access to long-distance trade routes, as this allowed for the continued import of metals to the site. Social complexity and interregional exchange did not, therefore, disappear in the later IA. Insularity may have been a feature of Beycesultan's society at this time, however. This is meant not in the sense of a lack of external contacts (as these are attested by the continued metals trade) but, rather, in terms of no longer having an outward-looking perspective. The wide range of external styles and influences evident from the previous phase narrows dramatically in Level Ia, and the decorative schemes used for ceramics return to having a distinctly local flavor. The only external Other that was represented in the community's material culture seems to have been Cilicia and the southeast. The community at Beycesultan would have had to maintain connections with this area to ensure continued access to metals, and it not only may have been familiar with southeastern material culture but may have viewed this connection favorably. This perhaps explains why the inhabitants there were comfortable enough with the southeastern connection to make use of items that evoked it during enactments of community. Given the lack of evidence both for this archaeological level and for the period more generally, it is impossible to fully explain the crystallization of a conscious community identity in IA Beycesultan. It may have been either due to the community coming together again in the face of an external threat or an expression of the community's autonomy and an inward-looking perspective. Until more is known about the wider regional dynamics of this period, the specific social dynamics at Beycesultan must unfortunately remain unexplained.

The Phrygian Level Very little information has been recorded on the Phrygian Level. All that is known is that a domestic unit uncovered in Trench Z on the south slope Page 135 →of the east mound had a number of Phrygian-style vessels on the floor. Using companies from Gordion, these vessels were identified as belonging to the eighth or early seventh centuries BC. According to Rodney Young, the director of the Gordion excavations of 1950–73, there are very close parallels from the early seventh-century destruction layer at Gordion (Mellaart and Murray 1995, 89). These included large storage vessels such as pithoi as well as decorated finewares, suggesting a range of social activities in this area, from dining to storage. However, only sixteen of these vessels have been published, none of which were registered in the official documentation. No information is recorded about the house in which these vessels were found or about any small finds that may have been in the same structure. In addition, nothing is known about any other contemporary structures at the site, and nothing can be said about the nature and extent of the site's occupation at the time. The remains of any contemporary constructions on the mounds themselves either would have been destroyed by later building activity or were ignored by the excavators as unrelated to their central research interests. As so little has been recorded concerning the Phrygian level, it is impossible to make any general claims about how people in Beycesultan during this period chose to represent external connections or articulate its identity. All that can be known is that sometime during the eighth or early seventh century, the inhabitants of Beycesultan had some contacts with Phrygia to the northeast and expressed some affinity with this area through their use of characteristically Phrygian

fineware styles (see, in particular, Cat. no. B5.07). During this period, the state of Phrygia was at the height of its powers, and Phrygian cultural influences were felt elsewhere in western Anatolia at Sardis and Daskyleion. It is therefore not surprising that some individuals within the community at Beycesultan would have used and appropriated Phrygian high-status styles, symbolically linking themselves to the established elites at Gordion. It is unfortunate that we can only speculate on what significance Phrygian styles would have had at Beycesultan. Would it have been the accepted style of material culture associated with elites and high status, the accepted style of material culture associated with middling status or status competition, or not an accepted style of material culture at all but a bold and conspicuous statement of unusual self-association and political allegiance? Unfortunately, the limits of the material available preclude any real understanding of the Phrygian pottery's social significance. Page 136 →

Conclusions: Communities Forged in Fear? LBA-IA Beycesultan The first LBA phase at Beycesultan begins around the end of the thirteenth century and start of the fourteenth. At this time, major construction works were undertaken that would shape the urban layout on a pattern that would last for over two hundred years. Stability, continuity, and a strongly local community identity were vital to the inhabitants of Beycesultan at this time. A sense of community was strong during this phase, and solidarity was maintained through commensality and the standardization of the object associated with commensality. This strong community identity was closely linked to a sense of locality and emplacement in the immediate neighborhood of the Upper Maeander valley. This local flavor was stressed in the forms of material culture used during enactments of community, despite political influences from the Hittite Empire to the east and the Arzawan confederacy to the west. The conscious expression of a specifically local community identity in the face of these integrating forces—one pulling eastward, the other westward—is suggestive of the community's independent self-definition. In the closing decades of the LBA and the first few of the IA, Beycesultan became an increasingly centralized and more rigidly hierarchical society, perhaps benefiting economically from the collapse of the larger states that had previously dominated the eastern Mediterranean. While maintaining its own local community identity and association with the Upper Maeander region, it also played up associations with the northwest, particularly the area around the Hermos valley. Preserving a sense of independence, then, Level II Beycesultan simultaneously articulated an identity that was western-oriented, at a time when this particular area seems to have been politically coming into its own. However, the sense of community and collective togetherness that is conspicuous in the previous phase is absent in Level II. Instead, hierarchy, rank, and distinctions within the community became more important than the solidarity of the community as a whole. Wealth and resources became increasingly concentrated within select households, and feasting both was no longer carried out on a large scale and no longer employed standardized forms of material culture. The stress on differentiation only increased in Level Ib. However, instead of a rigid hierarchy, this phase saw a much more fluid social structure characterized by social mobility and the active creation and contestation of status. Inequalities were no longer enshrined in the built Page 137 →environment, and status instead seems to have been negotiable. The competitive nature of the status quest during this time meant that no discernible sense of community identity could be identified in the archaeological record. Internal distinctions, rather than the distinction between the communal us and the external them, were of crucial interest at this time. Competing external affiliations also characterized the period, and different elements of Level Ib society stressed different external relationships. Links to these various areas could be played off one another and used, like other social strategies such as bodily adornment or competitive consumption, in the ongoing status competition of the time. In Level Ia, however, this instability seems to have evened out, and some sort of central social organization seems to have returned to Beycesultan. As well as centralized organization, there also seems to have been a sense of

corporate solidarity and togetherness during this phase, a sense conspicuously lacking in Levels II and Ib. A central community building, used for enactments of community that encouraged open and undifferentiated participation and that used standardized forms of material culture, would have been instrumental in forging this community identity. The community itself again appears to have been rooted in the immediate Upper Maeander locale. However, links with the southeast—an area that may have been the eastern Mediterranean's economic hub at the time—were also recognized. A gap in the evidence ensures that we cannot know what, if anything, went on at Beycesultan for a long period after the abandonment of Level Ia. But the limited evidence available shows that sometime in the eighth century BC, at least a section of the community was symbolically engaged with Phrygia to the northeast. While nothing further can be inferred, it is unsurprising that there were those at Beycesultan who used material culture to associate themselves with the dominant political power of the time. As in previous periods, the eighth-century people of Beycesultan were not isolated from other contemporary sites and societies; rather, they were closely involved in the wider developments of the time and profoundly influenced by events not just elsewhere in Anatolia but in the Mediterranean as a whole.

Community Identity in LBA-IA Beycesultan Social developments at Beycesultan are not always easy to infer from the various patterns, signs, and red herrings left to us in the material record. Page 138 →However, we can trace a social trajectory, which is both linked to wider developments elsewhere in Anatolia and the eastern Mediterranean and also specific to Beycesultan itself. Evidence for a conscious community identity at Beycesultan can be found at two specific points within the site's history, apparently at times of stress and external pressure. The first of these is during the LBA, when the settlement was suffering the unwelcome attentions of the Hittite Empire. This external pressure from an expansionist imperial state seems to have encouraged a collective mentality at Beycesultan, with the community banding together in opposition to the external threat, suppressing its internal differences, and playing up group solidarity and unity. The second of these points is during the later IA, a period of widespread instability across the eastern Mediterranean. At this time, the inhabitants of Beycesultan seem to have withdrawn inward somewhat, closing ranks against external influences and focusing again on their local area. The common thread between these two historical contexts is one of potential threat and uncertainty. Faced with either an imminent menace or a more generalized environment of insecurity, the people of Beycesultan cultivated a conscious sense of community identity, taking solace in solidarity during times of perceived danger. Community identity at Beycesultan therefore seems to be linked to fear. It is unsurprising that fear can be a crucial factor in communities coming together—within living memory, experiences of local societies in wartime bear witness to this tendency. Perhaps more interesting in the case of Beycesultan are the times when community identity did not crystallize, the times at which the inhabitants of the site did not feel sufficiently afraid to form a coherent community. Intriguingly, this seems to have been during the LBA-EIA transition and during the early years of the EIA. This period is usually thought of as a time of social transformation and political revolution. As described in chapter 6, the breakdown of states and empires at the end of the LBA was thought to usher in a period of turmoil and instability across the eastern Mediterranean, characterized traditionally as collapse and more recently as transformation. At Beycesultan, however, this period of change does not seem to have stimulated a sense of threat or fear and did not lead to the formation of community identity. Instead, the changes happening on a broader interregional scale seem, if anything, to have opened up new economic niches and political opportunities that groups within the Beycesultan population were quick to exploit. Status distinctions became dramatically more important during this period, as individuals, Page 139 →households, and groups sought to distinguish themselves ever more clearly from each other. This interpretation has important implications for our understanding of the eastern Mediterranean as a whole during the LBA-EIA transition. The experience of this transition was far from uniform across the eastern Mediterranean or even within regions such as Anatolia. At Beycesultan, it is marked by the removal of an external threat and a release from fear and simultaneously by an accompanying decline in community identity and a growing interest in status distinctions. Whatever may be said about the LBA-EIA transition elsewhere, in this one

settlement, it appears to have given people little cause for upset or concern.

Page 140 → CHAPTER EIGHT

Hierarchy and Community at Aphrodisias The Site and the Material Introduction Investigating community identity at Aphrodisias presents a very different set of challenges and opportunities from Beycesultan. Aphrodisias differs from Beycesultan both in terms of the type of archaeological material preserved and also in terms of the nature of the site itself. Most notably, Aphrodisias is well known as a Roman site, rather than a prehistoric or protohistoric one. However, the site's earlier levels were investigated in six seasons of excavation between 1966 and 1974 (fig. 13). These excavations have showed that Aphrodisias was settled as early as the Late Neolithic period (ca. 5800 BC) and occupied through the Chalcolithic, Bronze, and Iron Ages, although there seem to have been a number of gaps in occupation (Joukowsky 1986, 176). But while the material from the Early Bronze Age levels has recently been reassessed (Greaves 2009), the LBA and IA levels remain relatively poorly understood. Another major difference between Aphrodisias and Beycesultan is its location. The site itself lies in the fertile valley of the Dandalus, a tributary of the Maeander branching off to the south before the river reaches the Denizli plain. The valley is enclosed to the east and the west by mountain ranges, with the Baba Da range behind Aphrodisias to the east and the Karinçali Da i across the river from Aphrodisias and beyond the modern town of Karacasu to the west, and it leads, for all intents and purposes, to a “dead end” in terms of external routes to the south. At its southern extremity, the Dandalus valley is connected to the Tabai plain by a single hilly pass, itself an isolated area due to its topography (Robert 1954, Page 141 →18-22, 38-39). While the Dandalus valley therefore mediates the Tabai plain's connections to major trade and communication routes, its only real point of access to these routes was to the Maeander valley in the north. Aphrodisias was therefore not on a major route or thoroughfare—it was essentially located on a “road to nowhere.” Direct access from the valley to the wellconnected Denizli plain would have been extremely difficult, as it would have required crossing the Baba Dag range, with peaks as high as 2,421 meters (Robert 1954, 17). In stark contrast to the area immediately around Beycesultan, the Dandalus valley shows relatively little sign of occupation in either the prehistoric or classical period, as no höyük mounds other than those at Page 142 →Aphrodisias, and only a few later sites are known. Indeed, the closest classical period sites are on the Tabai plain (Heracleia, Tabai, and Apollonia), tucked behind the Karincali mountains (Plarasa), or at the junction of the Maeander and the Dandalus (Antiocheia). Mellaart's surveys of the wider region reported only one mound with signs of activity from the Bronze or Iron Ages—that of Sarayköy. Even Sarayköy is fairly far from Aphrodisias, lying about forty kilometers away, where the Dandalus meets the Maeander. ükrü Tül's 1980-81 survey of the Maeander valley highlighted neither classical nor prehistoric sites in the Dandalus valley (Tül 1986), and a brief survey by the Aphrodisias excavation team itself in 1982 discovered no sites with preclassical remains (Leurquin 1986, 734). Aphrodisias, then, seems to have lain in a largely unpopulated landscape throughout much of antiquity and seems to have been one of very few LBA and IA settlements in the Dandalus valley. The site currently covers a large flat area on the east side of the valley, but early occupation is thought to have been primarily on two conical mounds, the Acropolis mound and the Pekmez mound. The whole area is overlaid with Hellenistic and Roman remains whose construction has badly disturbed earlier archaeological levels. The most obvious example of this is the Hellenistic theater that was cut straight into the side of the Acropolis hill. Both of the mounds have been further disturbed by more modern activity. The village of Geyre was built amid the ruins of Aphrodisias. In 1979, the Turkish government forcibly relocated Geyre one kilometer to the west for conservation reasons, and the slopes of both the Acropolis and the Pekmez are covered in recent rubbish pits and modern house foundations (Joukowsky 1986, 20). Consequently, the LBA and IA levels have not been well

preserved, especially in comparison to the prehistoric Early Bronze and Chalcolithic levels, which were further from the surface. The remains of LBA and IA levels have been found only on the Acropolis mound (fig. 16), which, at a maximum height of twenty-four meters from the plain, has been eroded less than the Pekmez mound, which rises to a maximum height of thirteen meters (Joukowsky 1986, 19). Aphrodisias is a good comparison for Beycesultan, because of both their similarities and their differences. The sites were occupied in similar periods, and they are among the few examples of excavated inland sites in western Anatolia, other notable exceptions being Sardis and Daskyleion. They also share a number of ceramic styles and cultural traits. Aphrodisias is often seen as being a second-or third-order site in the general culture area of which Beycesultan is thought to be the type site (Joukowsky 1986, 162, 468). However, it will be informative to trace where the similarities and differences really lie between each community's social developments and their responses to similar historical situations. In addition, the two sites are very different in terms of the types of society that inhabited them and the type of social life that would have been experienced there. Page 143 → Page 144 → The topographical nature of the sites is an important point of contrast, as early Aphrodisias seems to have covered two smallish but separate höyüks as well as part of the flat plain between them, instead of being focused on a large twin höyük with occupation spreading out onto the plain around as at Beycesultan. While the settlement at Beycesultan had an obvious spatial focus, being clustered around its central twin mound, Aphrodisias may have been more of a dispersed community, with occupation perhaps in several discrete clusters. Similarly, while Beycesultan seems to have been a central settlement in a densely occupied area and the largest site in a full archaeological landscape, occupation in the Dandalus valley seems to have been more dispersed in general, with few settlements in Aphrodisias's immediate area and with no obvious settlement hierarchy among those that are known. While Beycesultan was located at the mouth of a pass leading up to the Anatolian plateau and on an important line of communication linking the plateau and the Aegean coast via the upper and lower Maeander valleys, Aphrodisias was situated on a dead end as far as long-distance routes were concerned. The community at Aphrodisias was therefore of a qualitatively different type than that at Beycesultan, and its experiences and relationships with the external would have also been very different. Practicalities The preclassical remains have been published as fully as possible by Joukowsky in her 1986 monograph (1986; earlier excavation reports include Erim 1976, Kadish and Erim 1969, and Marchese 1976). The recording and recovery strategies of the Aphrodisias excavations were more thorough than those at Beycesultan, as archaeological methods and practice had changed over the twenty years intervening between the two projects. Joukowsky's monograph therefore incorporates more information from the site diaries; a painstaking reconstruction of the stratigraphy, even where it was disturbed; computer analysis of ceramic data, including proportions of shapes, wares, and decorative features in the assemblage over time, using contemporary statistical methods; and a complete catalog of the known stored small finds and diagnostic ceramics (see Page 145 →appendixes C and D in the present study). Many of the finds from Joukowsky's catalog, however, including most of the objects cataloged from the LBA and IA levels, have unfortunately since been destroyed in a fire in one of the on-site storehouses, so this primary material can no longer be consulted (Bert Smith, personal communication).1 The ceramics in the catalog are largely sherds, but each entry represents a complete vessel, as sherds potentially from the same vessel were recorded together. However, this means that it is relatively difficult to draw much in the way of conclusions about the shapes of the vessels, in terms of either their functional category or whether their shapes would have been thought of as characteristic of any particular areas. Information is perhaps more in evidence for the former than for the latter, because the general type of shape is usually easier to assess than the specific form. While it may be possible to say a certain sherd came from a bowl, for example, it may always not be possible to infer what the whole profile of that bowl may have been. Therefore, while some of the following analysis includes discussion of the functional categories of pottery from Aphrodisias, there will be little reference

made to how vessel shape may have recalled external associations. Exceptions to this involve individual vessels of which enough remains to identify a shape characteristic of a specific region, such as the Ionian cups and lydion found in Level IIID. The high level of recovery and documentation makes the ceramic assemblage from Aphrodisias a much more rounded and full sample than that of Beycesultan, which was subject to a number of explicitly selective biases. However, the absolute numbers of objects that can be assigned to each specific phase at Aphrodisias are relatively few, and the size of the Aphrodisias assemblages is much smaller than those available for Beycesultan. Although the Aphrodisias assemblages are more representative of the areas excavated than those of Beycesultan, the total area excavated at Aphrodisias is much more limited. Sherds belonging to a total of 473 vessels were published from six archaeological phases (appendix D). The material from Aphrodisias is therefore inappropriate for quantitative analysis, as the sample sizes for most phases are too small to be statistically reliable. As with Beycesultan, the use of percentages is only included here for some of the assemblages and should be taken as only a rough guide, rather than as being statistically significant. LBA and IA material was unearthed in three trenches on the Acropolis mound, Trenches 6, 8, and 9 (fig. 14). Each trench unfortunately contained material from slightly different time periods, due to the different taphonomic processes at work in each area. The area where Trench 6 was Page 146 →sunk seem to have been used as a cemetery area throughout the Bronze Age (Joukowsky 1986, 119-21) and contains the disturbed remains of pithos burials from the MBA and LBA as well as a single cremation from the late seventh or early sixth century BC (Benjamin 1986). The extension of Trench 6 uncovered more later IA remains, including substantial quantities of painted fineware pottery in a distinctively Lydian style (Mierse 1986). The areas in Trenches 8 and 9 both appear to have been general work and domestic areas, containing architectural remains from the final stages of the LBA and the IA. There are therefore a total of six distinct phases at LBA-IA Aphrodisias: three from Trench 8 (Levels 4.III, 4.II, and 4.I), two from Trench 9 (Levels 5.II and 5.I), and the final Lydian phase of the Trench 6 extension (Level IIID). Although some remains from the cemetery area in the main trench, Trench 6, would have been contemporary with the Trench 8 phases, the disturbed nature of the stratigraphy makes it impossible to date individual artifacts, and therefore none of the LBA Trench 6 material has been included in this case study. The single cremation from the later IA period in Level 6, however, can be said, on stylistic grounds, to date to the phase of the Trench 6 extension and is considered alongside this material. The total area uncovered for each phase is relatively small, with Trench 8 occupying not more than seventy square meters, Trench 9 not more than eighty, and Trench 6 not more than sixty, while the extension occupied forty-nine square meters. This is a very limited window onto early Aphrodisias, offering evidence only for a small part of what must have been a much larger settlement. We therefore have access to a smaller sample of the overall settlement at Aphrodisias than was available for Beycesultan, and it is impossible to extrapolate much information about the population as a whole. The section of the settlement that can be examined for any given phase can thus only offer us a skewed view—portraying the perceptions of only one faction or one household within the wider settlement group. Nonetheless, although we can only examine the construction of identities carried out by whatever social groups were active in the excavated area, these identities could only have been constructed with respect to the community as a whole. They would therefore represent one part of the wider social dynamics at the site and are still significant. However, although examining one part of the whole is certainly a valuable endeavor, it should nonetheless be remembered that the remains of Aphrodisias only show us one piece of the overall picture. There are therefore many significant differences between the two case Page 147 →studies considered in this book. This is true both of the kinds of material available for analysis and of the types of interpretation that can be inferred from this material. Whereas the remains of Beycesultan allowed for a wide-angle view at a relatively low resolution in terms of detail, the Aphrodisias material offers a narrow view but at a much higher resolution. Investigating the evidence for community identity and its social rationale in this context will be quite different from investigating it at Beycesultan.

Chronology Joukowsky suggests, on the basis of stratigraphy, that the three building phases of Trench 8 (Levels 4.III, 4.II, and 4.I, from earliest to latest) and the two phases of Trench 9 (Levels 5.II and 5.I, from earlier to later) should be considered as roughly consecutive (1986, 174-75). She is, however, reticent about the precise dating of these phases, tentatively suggesting that phases represented in Trench 8 belonged to the LBA and that those of Trench 9 belonged to the IA, while openly admitting that her chronological scheme is based on very limited evidence because of the lack of solid ceramic comparisons that can be made (162-63). This rough chronology can, however, be improved on and refined, especially in the light of new ceramic material from Kilise Tepe and Sardis as well as the new chronology I have suggested for Beycesultan. The relationship with Beycesultan is of particular importance, as it is the excavated site closest to Aphrodisias in terms of both geographical distance and ceramic trends. While the absolute dating of the LBA and IA archaeological levels of both sites remains obscure, it is possible to place them in a relative chronology, so that comparisons can be drawn between each community's social development. The crucial phases for dating at Aphrodisias are Levels 4.III and 4.II, as developments that have the clearest external parallels occur in these building phases. Type A1 lustrous wares were popular at Aphrodisias in Levels 4.III and 4.II but suffered a dramatic drop in popularity in either Level 4.I or Level 5.II, comprising only a relatively small proportion of the ceramic assemblage from these levels onward (table 7). Similarly, Type B1, an innovation on the Type A1 lustrous wash, constitutes 57.3 percent of the total assemblage in Level 4.III but drops in popularity dramatically in Level 4.II, to 8.5 percent, before falling out of the repertoire completely until the Lydian period. The exact timing of this decline in popularity of both types of lustrous ware is uncertain, however, as the size of the sample from Level 4.I is too small to draw firm conclusions. However, the period of lustrous wares' main use at Aphrodisias, in Levels 4.III and 4.II, Page 148 →could be seen as very broadly contemporary with their popularity elsewhere in western Anatolia. While the aesthetic of lustrous washes is known from the LBA in the Hermos river valley, its use peaks at Beycesultan in Level II around the time of the LBA-IA transition (although the lustrous wash aesthetic appears at Beycesultan more on Types A1 and A2, rather than on Types A1 and B1 as at Aphrodisias). Located geographically between these two areas, therefore, the Aphrodisias of Levels 4.III and 4.II can probably be considered as belonging to a roughly similar time frame, from the LBA, and as broadly dating to the time of the LBA-IA transition. This general time frame is made more specific by the appearance in Level 4.II of a vessel that bears striking similarities with the Knobbed Ware Level VIIb at Troy. This bowl (Cat. no. A2.28) has pronounced grooves under the rim and knobbed protrusions as decoration (fig. 17). Although this vessel does not adhere closely to Knobbed Ware forms—in particular, the knobs curve upward in a fashion not usual in true Knobbed Ware vessels—it does appear that it was inspired by the general aesthetic (Carolyn Chabot Aslan, personal communication). Taking the current dating of the appearance of Knobbed Ware at Troy to ca. 1130-1050 BC (Level VIIb2; Becks 2003, 49) and assuming that it would have taken some time for the stylistic influences to travel southward from the Troad to the Dandalus valley, this would place Level 4.II some time in the late twelfth or early eleventh century and firmly in the IA itself. It seems, therefore, that Level 4.III may have covered the time of the LBA-IA transition (ca. 1200 BC), while Level 4.II followed it, in the latter part of the twelfth century, perhaps continuing into the eleventh. This late Aphrodisias chronology is not as unlikely as it might initially seem. It would place the popularity of Type B1 decoration, the distinctly Aphrodisian style of red or orange painted decoration on silver or gold lustrous washes, in a similar temporal context to the tightly painted linear decoration of the Submycenaean and late LH IIIC styles of the twelfth-century Aegean (Mountjoy 1993, 114-18), which are known to have been used at this time at Bademgedigi Tepe on the Aegean coast (Meriç and Mountjoy 2002; Meriç 2003). The Aphrodisias linear painted pottery is densely decorated with roughly painted bands, wavy lines, and diagonal hatching and shows a preference for warm colors for painting—primarily reds, browns, and oranges—on a lustrous background. This compares with the dense but rough painted decoration of Submycenaean and LH IIIC late wares, which also tend to feature warm colors of paint. The styles are clearly different, and Aphrodisias linear painted decoration Page 149 →cannot be said to derive directly from Aegean styles, but these general aesthetic schemes are comparable and could have plausibly arisen in roughly contemporary societies, which lends some support to my late

chronology for Aphrodisias. Broadly contemporary with, but probably slightly later in date than, the levels from Beycesultan considered in this book, Aphrodisias was a different type of settlement. It would have housed a more spatially dispersed population, located in a relatively isolated area and within a sparsely populated landscape. The two case studies offer the perspectives of two very different settlements in inland southwestern Anatolia, embarking on two different social trajectories and showing two different sets of responses to the events and circumstances of the LBA and IA. Beycesultan and Aphrodisias also had very different patterns of external interaction due to their respective locations, coalescing at different times around a collective sense of community identity and constructing their internal identities using different representations of the external Other.

Level 4.III The Remains of Level 4.III The remains of Level 4.III at Aphrodisias consist of what appears to have been a primarily domestic building complex, constructed from mudbrick on stone foundations (Joukowsky 1986, 149). Foundations from the Roman period have disrupted the remains in this area, meaning that it is not possible to draw a complete plan of the domestic unit. However, there do appear to have been a number of distinct spaces within the complex (fig. 15). The main room within the complex measured six by four meters and contained a stone-lined hearth with ash and charcoal on the floor around it and several crescent-shaped loomweights nearby. The working of textiles within the building complex is also suggested by the presence of several spindle whorls. These remains suggest that this main room was a multipurpose craftworking area, where activities such as textile weaving were undertaken. A smaller room to the south may have been used for storage, as the remains of a pithos were found there. A substantial number of stone tools were also found in the building complex, most of which are cutting and scraping tools and are likely to have been used in some kind of craft activity. Page 150 → It does not appear that these tools of the hearth were primarily used for food preparation, because of the small numbers of ceramics found that could potentially have been used for cooking. Only two vessels could possibly have been used in cooking food, a conspicuously small proportion of the overall assemblage when compared with other phases (table 8). This is especially notable given the high standards of collection and recording employed at Aphrodisias. If there had been more evidence for food preparation Page 151 →in this structure, it would most likely have been recorded. In addition, it is not certain that the two vessels that were found would necessarily have been used for heating foodstuffs. Instead, they may well have been used to heat substances used in the craftworking and textile production. In addition to the craft activities that seem to have taken place in this building, group dining or feasting also seems to have featured. A large number of vessels found here were dining shapes, comparable to the numbers of vessels found in the dining clusters at Beycesultan (cf. tables 2 and 5). This concentration of dining vessels within one structure clearly points to commensality. In addition, the majority of the sherds classed as having “unclear” function are likely also to have been dining vessels, as they were of identical fabric and decoration to the known tablewares (table 8). If this is indeed the case, this would imply even larger numbers of diners. Little can be inferred about the household that used this multipurpose space. (In this context, I use the term household to refer to the group or groups of people using the domestic unit excavated. The term is therefore applied only loosely and implies merely a social group focused on a particular architectural space, which may or may not be linked to a particular biological family or kin group.) The evidence available tells us nothing of the internal makeup of this household, suggesting only that it was engaged in two main activities: craftworking and group dining. But perhaps this lack of information about the nature or composition of the household is in itself significant. In this phase, there are no clear signs of internal status differentiation or the articulation of intrahousehold identities, as there were no items found that were clearly intended for personal status display, such

as jewelry or ornate weaponry. Spatial differentiation or the clear demarcation of “quarters” for different social roles is also absent. Level 4.III Enactments of Community It does not seem, therefore, that the household here was at pains to emphasize divisions within their ranks. Nor, however, is there much sign of household solidarity per se. Instead, it seems that the people who used the excavated structure were most crucially interested in reaching out beyond the immediate household to the wider population of Aphrodisias. The high proportion of dining vessels in the ceramic assemblage suggests that feasting or at least commensality took place within this structure (table 8). The absolute numbers of vessels involved imply the involvement of not just the immediate household but a reasonable proportion of the wider population of the site. Page 152 → Feasting therefore took place within the excavated area of Aphrodisias, providing an appropriate venue for enactments of community. As in the first phase at LBA Beycesultan, the ceramic vessels used in LBA Aphrodisias were also very standardized (table 7). While the dining events staged in this structure during the Level 4.III phase at Aphrodisias employed a relatively wide range of different styles, it did so only in very small numbers. Only single examples have been found, for instance, of Type D black and grey wares or Type C1 Upper Maeander coppery lustrous wares. Instead, the vast majority of dining vessels used in this phase belonged to a single type—Type B1, linear painted decoration on a lustrous background, which constituted 57.3 percent of the overall assemblage. The next most common style, Type A1, is also closely related to the dominant Type B1, as Type B1 is a local innovation on the plain lustrous style of the Hermos valley region. The overwhelming popularity of Type B1 and the fact that the second most popular style is the closely related Type A1 show the high level of standardization in the pottery used for dining in the Level 4.III structure. The gathering of substantial numbers of people to eat together using standardized dining vessels suggests that an ideology of togetherness was being promoted by those using this structure in Level 4.III Aphrodisias. This enactment of community through standardization and commensality is closely paralleled by that seen at Level III Beycesultan (see chapter 7). Although little can be said about the rest of the population at this time, it does appear that at least one section of the site's inhabitants were constructing a communal group identity here. However, the numbers of diners involved are comparable to those involved at Beycesultan, so it is possible that the group togetherness being enacted in this structure would have contributed to a sense of overall community identity. Given that the Aphrodisias site was neither particularly large nor particularly important at this time (the close of the LBA), these diners may well have represented a reasonable section of the wider community. It is therefore possible that the evidence uncovered by chance in this particular area does suggest that enactments of community were being carried out in Level 4.III Aphrodisias. Level 4.III Representations of the External Other Unlike Beycesultan, Aphrodisias does not seem to have had close or frequent contacts with the outside world. The most notable imported items come in the form of stone objects (table 9). These include a burin made Page 153 →from garnet (Cat. no. A1-1), some quartz objects (Cat. nos. A1-4, A1-12, A1-17), an obsidian piercer (Cat. no. A1-19), four smoothed marble pebbles (Cat. nos. A1-25 to A1-28), and a fragment of red marble (Cat. no. A1-29). It is almost impossible to ascertain the original provenance for these stone objects, with the exception of the obsidian piercer. Aphrodisias is known to have sourced obsidian from both Melos and Cappadocia during the Chalcolithic period (Blackman 1986), but it seems likely that the piece from this phase originated from Cappadocia, as it is black translucent while Melian obsidian is known for its greenish tinge. However, this object did not necessarily evoke associations with central Anatolia, even though it probably was imported from there. The piercer does not seem to have been used during enactments of community or at other socially significant events and was deposited in exactly the same fashion as the other stone tools.

Crescent-shaped loomweights are another class of object that might imply external contacts but that may not have recalled associations of these contacts. This type of loomweight appears at Beycesultan and Kusura as well as at Aphrodisias, perhaps indicating some common practices in textile production over the inland southwestern Anatolian area (see chapter 6). However, in Level 4.III Aphrodisias at least, these items do not seem to have been intended for display. There is a total lack of decoration in any form on the examples from this phase, contrasting sharply with the slips, stamps, pinpricks, or incised decoration found on examples from Beycesultan, Kusura, and even Level 5.II of Aphrodisias itself. This suggests that at this time in Aphrodisias, crescent-shaped loomweights were not treated as being particularly significant objects and may not, therefore, have recalled associations of the external Other for the inhabitants of the site. It is in the decoration of ceramic finewares that the most obvious references are made to external Others. Foremost among these is the preference for silver and gold lustrous washes, either plain, as in Type A1, or with painted linear decoration, as in Type B1. The lustrous style, generally associated with areas directly to the north of Aphrodisias around the Hermos river valley, is likely to have recalled this area for the inhabitants of Aphrodisias. However, the preference for Type B1 over Type A1 indicates an inclination toward the local innovation rather than the original external style. This appropriation and reworking of an external style suggests that while connections with the northwest were acknowledged, they were mediated by a more local sense of identity. In this respect, the inhabitants of Level 4.III Aphrodisias appear to be doing something similar to those Page 154 →of Level II Beycesultan, adopting the same external style, adapting it to suit local tastes, and producing a visually powerful composite style that can only be thought of as characteristically local. The local specificity of these adaptations is evident from the almost total absence of Type A2 wares at Aphrodisias, as by the very small number of Type B1 decorated vessels at Beycesultan. Each of these sites preferred its own particular local version of the Hermos valley style. It is significant that links with the area around Beycesultan and the Çivril valley are not emphasized at Aphrodisias, with only one characteristic Beycesultan Type A2 vessel appearing in the assemblage. The same can be said for several external Others at Level 4.III Aphrodisias. Some contact with the southeast can be assumed from the presence of some Type E vessels, and some connection to the far north can be inferred from the Type D black slipped bowl. The presence of these types indicates an awareness at Aphrodisias of wider Anatolian stylistic trends but these broad influences were simply not popular. Both the household using this structure and the wider community that this household seems to have engaged must have specifically chosen not to incorporate external influences in their ceramic repertoire. Level 4.III Community Identity and Its Social Rationale From the evidence uncovered here, a cohesive group identity was enacted in Level 4.III Aphrodisias, a group identity that may well have roughly equated with an overall sense of community identity because of the small scale of the site. This sense of togetherness was forged around the end of the LBA and in the early decades of the EIA, roughly contemporary with Level II Beycesultan. This was a time when the established powers of the LBA were in decline, with the collapse of the Mycenaean palatial system and the end of the Hittite Empire; when the international political order was being recast; and when new powers were emerging, most notably at Mira in the Hermos area and Tarhuntassa in Cilicia (see fig. 2 in chapter 6). Beycesultan's response to this wider historical situation was to lose its previous sense of community; no longer under threat from external aggressors, the inhabitants of Beycesultan stressed social differentiation and hierarchy, distinguishing themselves from each other as much as possible, as they no longer needed to bond together in the face of a common enemy. The inhabitants of Aphrodisias, however, seem to have responded Page 155 →differently to the contemporary situation. They appear to have emphasized the close-knit and local nature of their community, favoring local styles of material culture and staging large-scale events to bring people together in enactments of community. The reasons for this different response are unclear. Perhaps Aphrodisias, as a smaller and lower-profile settlement, was actually under greater threat in this period of change and instability than it was during times of more stable political control. That a modest settlement should show a sense of community solidarity during a time of political upheaval is perhaps not overly surprising. Or perhaps Aphrodisias, set in the fertile Carian hills, relatively close to the Hermos region, was under direct threat from the expanding power of Mira nearby.

The dominance of decorated pottery of Types A1 and B1 should perhaps be seen in this context. It seems that most external Others were not embraced through statements made in material culture. These ignored or excluded Others included communities in the Troad, the Upper Maeander, and Cilicia. The only external Other that gained some form of recognition is the Hermos valley area, as represented by the popularity of both Type A1 and Type B1 styles of fineware decoration. Both of these styles are ultimately derived from the Hermos valley area, the supposed heartland of Mira (Hawkins 1998, 2002). In significant occasions invested with social importance, it is noteworthy that the community of Level 4.III Aphrodisias used material culture that exclusively evoked a nearby political power on the rise. This may well have been the result of pressure put on the inhabitants of Aphrodisias—pressure that encouraged them to band together as a community but that also encouraged the community to align itself with Mira. We may never know whether this alignment was because of the external imposition from Mira or due to a strategic initiative on the part of the Aphrodisians themselves. The preference for the local innovation on the lustrous style can be interpreted both ways. It could be the attempt of a local population to accommodate themselves to an unwanted external influence, but it could equally be a sign of the Aphrodisians' easy familiarity with Hermos valley styles and of the extent to which they felt comfortable with them. Either way, it is impossible to be sure from the material available. Fortunately, the remains of this phase tell us a great deal about some of the things happening at Aphrodisias at the time. Unfortunately, however, they only tell us about some of the things happening at the time. To understand more about the full range of social dynamics, more excavation would be needed. Page 156 →

Level 4.II The Remains of Level 4.II In Level 4.II, another stone structure, the plan of which is again obscure, was built over the remains of the domestic complex of Level 4.III (fig. 16; Joukowsky 1986, 149-50). No cooking installations were found in this structure, but the broadly domestic character of this building is suggested by the range of finds discovered within it. The ceramic assemblage includes shapes fulfilling all the various domestic functions, including cooking, storage, and dining. The small finds once again include objects associated with small-scale textile manufacture and a number of everyday stone tools. While the building generally appears to have been a domestic unit occupied by a single household, there is an unusually high proportion of large storage vessels within the ceramic assemblage, including the remains of at least eleven separate pithoi (see table 8). The storage capacity represented by these pithoi would have been substantial and more than necessary for strictly household purposes. It is possible that this building served as a central food storage facility for individuals beyond the immediate household or that the household itself controlled substantial resources. In either case, centralizing social structures are implied by the existence of such a storage facility, as are established social roles and status differentiation. Status differentiation is also suggested by the presence of objects potentially used to signal prestige, including two serpentine axes (Cat. nos. A2-17, A2-18), a serpentine adze (Cat. no. A2-16), and two perforated shells that were used as jewelry (Cat. nos. A2-41, A2-42). The association of prestige items such as these with a central food storage facility might suggest that the household did indeed control significant resources and was part of a relatively established elite group. Toward the end of the building's occupation or at some stage after its abandonment but before the construction of the next phase, four large pits, containing charcoal, ash, and ceramics, were dug into the floor, and soil was packed hard into the room (Joukowsky 1986, 149). The meaning of this act remains uncertain, but the deposition of burnt matter and the packing of the room suggests some ritual for the intentional “ending” of the building and the deliberate closing of this occupation phase. Page 157 →

Level 4.II Enactments of Community The material available does not show any evidence for large-scale enactments of community for Level 4.II Aphrodisias as it did for Level 4.III. This is perhaps to be expected, given the domestic nature of the structure excavated. Large-scale events such as feasts and ritual activities are not likely to have taken place in such a restricted setting. Enactments of community, therefore, may well have taken place elsewhere in the settlement. Page 158 → The nature of the finds, however, suggests that there may have been less communal spirit during the Level 4.II phase than during the phase that preceded it. There is a new concern evident for individual status and differentiation, in the form of privately owned exotic goods and jewelry. The centralized control of resources suggested by the remarkable storage capacity is also significant of differentiation between households and greater social stratification. In such an environment, it appears that parts of the population, at least, were deeply concerned with distinguishing themselves from others at Aphrodisias. This, more than anything else, suggests that a sense of coherent community identity may have been lacking at Level 4.II Aphrodisias. However, the limited nature of the material means that we cannot be completely sure of this. Level 4.II Representations of the External Other Although there is no evidence for a community identity at Level 4.III Aphrodisias (indeed, there is some evidence against it), there are a few signs of how Aphrodisias's inhabitants comprehended their external contacts. The shells found in this phase are unfortunately of limited value in this respect. One of the shells is a marine species and could not have been found in the immediate vicinity of the Dandalus valley. Although this find might imply that the site was incorporated, to some extent, in long-distance networks of trade, the single shell may equally have been curated from a previous period. Although it was an item that would have signaled individual prestige, it is unlikely to have had any clear associations with any particular external Other. The ceramic assemblage, however, does suggest some representation of external Others. The assemblage is of a qualitatively different type than that of the previous phase. It includes many more coarse wares and undecorated vessels and shows a range of domestic activities being conducted within the household, rather than the previous phase's focus on feasting. Even accounting for this, there are some major changes in the apparent decorative preferences. The popularity of the local innovation Type B1 finewares dramatically decreased, while that of the plain lustrous Type A1 rose. Instead of using the local style preferred in the previous phase, the household of Level 4.II therefore engaged directly with the lustrous fineware aesthetic of the Hermos valley to the immediate northwest, without adding any local alterations to the style. In this household, then, the external Other now became the dominant insider. The Hermos aesthetic had taken over. Page 159 → One notable vessel is a large bowl with knobbed and ridged decoration, made of uneven coarse ware (fig. 17; Cat. no. A2.28). It bears many resemblances to the Knobbed Ware found in Level VIIb at Troy, which has in itself been linked to the ceramics of Bulgaria and Thrace (Becks 2003, 49; Blegen et al. 1958, 158). The style of the vessel is without parallel at Aphrodisias, and it would have been conspicuously different from other vessels used for dining. With a rim diameter of over forty centimeters, it is an especially large bowl and is likely to have been used at communal dining events where its unusual style would have been on display to individuals beyond the immediate household. But while the bowl was clearly different from local Aphrodisias styles, it nonetheless seems to have been locally produced, as it is made of local reddish yellow clay rather than the dark clays usually associated with the style in the Troad. So while this vessel was clearly influenced by a distinctive northern ceramic style, it was unlikely to have been an actual import from the far north. Level 4.II in Its Historical Context The lack of a clear community identity at Level 4.II Aphrodisias can be seen in its broader historical context. The

contemporary phases at Beycesultan (around the Level II-Level Ib transition) also show a comparable absence of community spirit and a comparable concern for individual status and internal social differentiation. In the case of Beycesultan, the removal of external threats and the guaranteed independence of the community may have allowed for this focus on internal dynamics. At Aphrodisias, however, the reasons for the breakup of its community identity may be somewhat different. The dominance of Type A1 styles of decoration in the fineware ceramic assemblage may give us some clues. At Beycesultan, the disintegration Page 160 →of community identity was accompanied by a widening of the geographical perspective and the strategic evoking of a wide range of different external Others; at Aphrodisias, the decline of community identity and the rise of individual differentiation was accompanied by the dominance of a single external Other—Mira. Aphrodisias may have come more directly under Mira's influence at this time, which may have limited the range of external Others that Aphrodisians could comfortably allude to in their material culture. While the standardization of Level 4.III Aphrodisias evidently came from choice (a narrow range of styles being preferred out of a wide selection of available styles), the standardization of Level 4.II Aphrodisias is altogether more severe, perhaps resulting from obligation rather than choice. The increasing influence of Mira may also account for the decline of community identity at Aphrodisias. The increased importance of personal status and the introduction of social hierarchy may be side effects of a new political order in the area. Intracommunity differentiation would certainly become more important as individuals and households competed for a role in this new order. This speculation, however, cannot be confirmed from the evidence available. Plausible explanations for the trends seen here can only be offered, rather than being fully substantiated.

Level 4.I The Remains of Level 4.I The structure in Level 4.I of this area is again built of mudbrick over stone foundations and again cannot have its plan reconstructed due to later cuttings (fig. 18). It is possible that this phase ended in a burnt destruction, as charcoal and ash covered the whole trench area (Joukowsky 1986, 150). There are signs that the individuals using the area in the Level 4.I phase were deeply concerned with personal status display, as several objects associated with personal bodily adornment were found, including a schist pendant (Cat. no. A3-45) and four pierced shells that could have been strung (Cat. nos. A4-57 to A4-60). A zoomorphic figurine made from chert was also found in this phase, but its condition was too fragmentary to allow for stylistic classification or for any parallels to be drawn (Cat. no. A3-44). The area does therefore seem to have been used by some individuals who invested their resources in material expressions of status. The nature of the building complex, however, and the social use of Page 161 →the area in general are still uncertain. The ceramic vessels found in this phase generally imply domestic activity, including cooking, dining, and storage. However, the evidence from these is weak at best, as the number of ceramic vessels found in this phase was extremely small. This is significant, because it suggests that the structure was not used primarily as a domestic house but that domestic-type activities may have occurred occasionally in the area. It seems that the area was primarily used for specialist craft production and perhaps for the production of high-status elite items. Among these elite items are several fragments of glass vessels and a lump of glass slag (Cat. nos. A3-51 to A3-55). These remains imply that glassworking, a highly specialized form of craft activity, may have occurred Page 162 →in this area. Another, perhaps more tenuous piece of evidence for the production of high-status goods in this area is the remains of a crushed murex shell (Cat. no. A3-61). This find might suggest the processing of murex for the production of purple dye, although the small amount of crushed shell found does not present particularly strong evidence for this. The presence of these two exotic and high-status materials in this area is significant. First, it suggests that at least some of Aphrodisias's inhabitants enjoyed considerable wealth and prosperity at this time, as indicated not only by these finds but also by the jewelry items mentioned earlier.

Second, it implies access to long-distance trade routes. Glass, murex, and the shell beads must have been imported to Aphrodisias from regions well beyond the Maeander river valley. Given that Aphrodisias was not located on a potential thoroughfare, the evidence for such external contacts is significant. It implies that the settlement of Aphrodisias had enough economic pull to have become a trading destination in its own right, as the site's location did not make it a convenient nodal point for exchange networks. Finally, in addition to wealth and access to trade routes, the presence of highly skilled individuals within the site's population may perhaps be inferred. The production of glass (and, more tenuously, of purple murex dye) would have required a high level of technical knowledge. Level 4.I Enactments of Community As with the previous level, Level 4.II, there are no signs of enactments of community or of the conscious construction of group identities in Level 4.I. Of course, the remains are focused in an area where we might not expect to see identities performed through social practice. A location used primarily for craft production is unlikely to have also been used as a venue for the creation and negotiation of large-scale group identities. Community identity could well have been enacted at other, more suitable locations within the site. However, as in the previous phase, some of the material found in Level 4.I suggests that internal stratification within the population was a great concern at Aphrodisias, making it less likely that a sense of collective community identity would have been expressed. The items of jewelry and the presence of high-status goods such as the glass vessels suggest that social differentiation was important at this time. If the interpretation of this area as a craftworking area for the production of prestige items is correct, it might imply that there was a reasonable level of demand for Page 163 →such items as well. Ranking, individual distinction, and personal identities therefore seem to have been a concern for the inhabitants of Level 4.I Aphrodisias, meaning that community identity may not have been high on their agenda. Level 4.I Representations of the External Other The remains of Level 4.I show Aphrodisias in a different light from previous phases. At this time, the inhabitants of Aphrodisias seem to have had greater access to imported objects and exotic prestige goods. The practical range of external contacts available to Aphrodisias, therefore, does seem to have increased. Around this time, Aphrodisias also caught up with developments that had already occurred in several other areas of Anatolia. Around the end of the LBA and the start of the EIA, a change is observed in the dining assemblages of sites in both the Troad and central Anatolia, where carinated bowls were gradually replaced by dishes. This shifting preference, from the deep-holding containers of carinated bowls to the shallow-holding platter of the dish, implies a change in diet and dining practices as well as in ceramic production. At Troy, carinated bowls were popular in Level VI (Blegen, Caskey, and Rawson 1953, 45-46) but not in Level VIIa or VIIb (Blegen et al. 1958, 26-27, 160-61). Similarly, the dish replaced the carinated bowl in the dining assemblages of central Anatolia toward the end of the LBA, although flat platters with stepped profiles had previously been used in food preparation (Glatz 2006, 131-32). At Aphrodisias, the same change occurred, not around the end of the LBA and opening of the EIA, but well into the IA itself, in Level 4.I. In this phase, the dish first appears in the repertoire of vessel shapes, and the carinated bowl declined in popularity. At Aphrodisias, the carinated bowl eventually dropped out of the ceramic repertoire altogether in Level 5.I (table 8). Aphrodisias's delayed catching up with this pan-Anatolian trend in Level 4.I suggests that the site was not well connected before this phase and only at this point became more integrated into wider regional trade. The remains of Level 4.I Aphrodisias therefore demonstrate the expanded network of relationships with external Others that Aphrodisias was practically experiencing. However, the evidence available cannot tell us for sure whether they were also incorporating influences from a wider range of external Others into their material repertoire. None of the small finds or imported items are reminiscent of any particular external place or can be said to have recalled specific external associations. In addition, the Page 164 →sample size of the pottery assemblage is too small to make any meaningful generalizations about stylistic preferences. However, the

presence of two handles decorated in Type D styles may potentially point to some acknowledgment of northern connections. The evidence for this, however, is too tenuous to support any conclusions. Level 4.I in Its Historical Context Overall, there is no clear evidence in the excavated material either for community identity or for how external Others were represented in the material record. This may partly be a product of the very limited view we have of the community, restricted to only a small area that seems to have been used for specialist craft production and would not necessarily have been a venue for the enactment of community identity. But the excavated area does suggest that the situation at Aphrodisias had changed greatly from the previous phase. Aphrodisias was better integrated into long-distance networks of trade and exchange and also enjoyed a reasonable level of wealth and prosperity. The population of Aphrodisias at this time probably included highly skilled craftspeople and also individuals who commanded substantial resources and displayed their wealth through personal adornment, forming a market for elite goods such as glass vessels. The expansion of Aphrodisias's network of contacts seems to be something new. This newness is implied in the uptake of new dining and dietary practices that had been adopted in other Anatolian regions some time ago. But while it is possible to identify a change in Aphrodisias's fortunes, it is somewhat harder to explain this change. Perhaps the pressure exerted on Aphrodisias by Mira had waned at this point, leaving the inhabitants of the site freer to engage in a variety of different economic activities. Level 4.I was occupied well into the IA, and it is easily possible that the power of Mira, emerging at the end of the LBA, was by this time on the decline. Certainly, the inhabitants of the contemporary Level Ib at Beycesultan were no longer using Hermos valley styles for ceramics at this time and were beginning instead to adopt Type E pale and white wares reminiscent of Cyprus and the southeast. Such an explanation might also help to explain the observed lack of community identity during this phase. While it is certainly possible that enactments of community were carried out elsewhere on the site, the evidence suggests that individual identities and status differentiation were crucially important for the inhabitants of Level 4.I Aphrodisias. This makes it far less likely that there would have been a coherent sense of Page 165 →community identity among the site's inhabitants. The removal of an external threat at Level II Beycesultan might have been behind the decline of community identity at that site, and a similar turn of events following the decline of Mira may well have had similar effects at Aphrodisias.

Level 5.II The Remains of Level 5.II Stratigraphically, Level 5.II follows on from Level 4.I, but it was excavated in the adjoining area of Trench 9 (Joukowsky 1986, 151-53). An external area to the south of Wall B and two rooms to the north of Wall B and separated by Wall A are preserved. The east room contains a large hearth built of stone and mudbrick, to the immediate south of which were a number of pithos fragments and the remains of a crucible lying amid ash and charcoal. To the north of the hearth, two pits were dug into the room floor; buried there were animal bones, more pithos fragments, and ash. In the southeastern corner of the other room was an enigmatic stone-built circle, packed with clay, sand, and ash and with unknown function (fig. 19). Like in the previous phase, it seems that the excavated area was primarily used for craft production. The crucible and the hearth, along with several lumps of formless metal and slag discovered nearby, suggest that metalworking was carried out in the excavated area (table 9). Other industries are also suggested by the large quantities of stone tools found, including whetstones, grindstones, and hammerstones, and also by the substantial numbers of objects associated with the chipped stone industry, including blade cores and stone flakes, which were described as being scattered “throughout the entire area of the trench” (Marchese 1976, 400). In addition, another murex shell was found in this phase, perhaps implying that whatever activities involving murex shells occurred in the previous phase may have continued (Cat. no. A4-154).

In addition to craft production, the area also seems to have been used for the large-scale production of food. Although all functional categories are represented in the ceramic assemblage, vessels used for cooking made up an unusually large proportion (24.5 percent) of the overall assemblage (table 8). This is more than would be expected for the production and consumption of food on a domestic scale and suggests that more food was prepared in this area than was eventually consumed, with communal dining taking place in other areas of the site. Of the dining shapes found in Page 166 →this area, most are large open vessels appropriate for communal service, rather than for serving individual portions. The unusually large numbers of these vessels also support the interpretation of this area as a place where food was prepared but not consumed. Level 5.II Enactments of Community The evidence found here for the preparation of food on a large scale implies that feasting activities may have occurred at Level 5.II Aphrodisias. However, as the consumption of the food would have happened elsewhere, Page 167 →it is impossible to tell whether these feasts would have stressed group solidarity (through standardized material culture and undifferentiated use of space) or distinctions within the group (through different classes of material culture and hierarchical space). After all, feasts can be a venue for the articulation of individual identities just as easily as they can be arenas for the enactment of community identity. In the case of Level 5.II Aphrodisias, it may be more likely that feasts stressed differentiation rather than integration, given the nature of some of the other small finds uncovered. Several prestige items and exotic imports show that status display and individual identities were of some importance to the inhabitants of Aphrodisias at the time. These include a metal pendant in a figure-eight shape (Cat. no. A4-152), two polished serpentine axes (Cat. nos. A4-137, A4-138), and polished stone adzes of blue marble (Cat. no. A4-104) and serpentine (Cat. no. A4-136). While we cannot therefore be completely sure whether the communal dining at Level 5.II Aphrodisias would have generated a sense of collective community identity or would have emphasized intracommunity divisions, the latter seems more probable than the former. Level 5.II Representations of the External Other As in the preceding phase, the inhabitants of Level 5.II Aphrodisias clearly enjoyed access to long-distance trade routes and widespread external connections. The specialized manufacturing industries that were located in this area would have required the regular import of raw materials, including, most notably, iron ores. Once contact with the wider eastern Mediterranean had been established, it seems to have been relatively resilient. However, little can be said for how the inhabitants of the site represented these external contacts in the material record or understood them more generally. Again, as in the previous level, no clear representations of the external Other can be identified. Neither the type of area excavated nor the nature of the remains that were found lend themselves to representation of the external Other. It is possible that such representations were performed materially elsewhere at Aphrodisias, but nothing can be said about them here. Level 5.II in Its Historical Context The community's involvement in metalworking makes it likely that it had connections to the southeast, in the direction of Cilicia and ultimately Page 168 →Cyprus, a major player in the pan-Mediterranean metals trade of the eleventh century (Sherratt 1994; Snodgrass 1994). Around this time, the Level Ib community at Beycesultan was also engaged in trade with this area and had embraced some of the Type E pale and white slipped styles that are characteristic of it. It is perhaps significant that while Aphrodisias was most likely in regular contact with the southeast, it did not choose to recognize this contact in its material culture. It is possible that Aphrodisias's independence and prosperity, newly emergent in the previous level, Level 4.I, had grown further at this stage and that Aphrodisias did not feel the need to acknowledge its trading partners. In the very obscurest period of what is sometimes called the “Dark Ages,” it is very difficult to conclude much about how Aphrodisias's inhabitants represented their external relationships. That external contacts did occur is not questionable—it is how these contacts were understood and presented that is uncertain. Most of what can be

reliably inferred relates to the basic details about Aphrodisias's internal social dynamics. Aphrodisias was home to a complex and relatively prosperous society, which placed importance on individual status and social differentiation and may have articulated these through feasting activities.

Level 5.I The Remains of Level 5.I No building plan can be inferred for Level 5.I, although the excavations have uncovered a curved area of hardpacked plaster with a small amount of charcoal ash in it, on a compact foundation of pebbles (fig. 20; Joukowsky 1986, 153). That this area was part of a building is likely, as there were postholes sunk into it that would have been used to support a roof. The careful treatment of the plaster floor also implies that the area would have been roofed, and the absence of any permanent walls within the confines of the trench suggests that these are the remains of a large indoor space, the walls of which lie outside the excavated area. Such a room, with its careful floor treatment, is likely to have been a central place of some kind and to have had some kind of public or semipublic use. This interpretation is supported by the ceramics found in the phase, which are nearly completely all dining shapes (table 8). Because of the almost exclusive focus on dining shapes in this area, it seems likely Page 169 →that this large and ornately constructed room would have been used for the socially charged activity of communal consumption. The few small finds from this phase broadly support this view, as potentially high-status items were found, including two polished stone spindle whorls (one of which was made from black marble), which are significant because all spindle whorls from previous levels of the site were made from clay (Cat. nos. A5-3, A5-4). In addition, the level yielded two shells, which, as in previous phases, were likely to have been used for personal adornment (Cat. nos. A5-1, A5-2). The discovery of such items fits into the wider picture of this area as a space for social dining and an arena for social signaling and identity discourse, and the very limited number of small finds (only six) suggests a specialized and nondomestic, noncraftworking use for the area, focusing on public events and communal dining. Level 5.I Enactments of Community The building excavated in this area lends itself to the performance of social identities, particularly to the enactment of community identity. The Page 170 →large, open space is undifferentiated and would also have accommodated significant numbers of people. The size of the structure seems to have been considerable, and although there is uncertainty over the exact scale of the building, it is certain that this was a substantial hall. Indeed, it seems that group activities were staged here, which could well have acted to encourage community togetherness and solidarity. Dining shapes dominate the ceramic assemblage, suggesting that this area was used for feasts (table 8). As mentioned several times during this book, feasts can act as an arena for the competitive construction of individual identities and for status contests. In this instance, however, they seem to have had a less combative feel to them. As was the case with Level II Beycesultan and Level 4.III Aphrodisias, the dining assemblage at Level 5.I Aphrodisias was highly standardized. This would have discouraged the articulation of difference, playing down the distinctions within the commensal group. Instead, parity and shared experience would have been played up, and a sense of community identity would have been actively fostered. This sense of collectivity is likely to have related to a good proportion of the site's population. While the absolute numbers of dining vessels found are relatively small, this is most likely due to later disturbance of the remains (which also means that the full extent of the building cannot be traced); the sheer size of the building suggests larger gatherings. A conscious sense of community identity can therefore be inferred for the inhabitants of Level 5.I Aphrodisias. Level 5.I Representations of the External Other

It appears that the community of Level 5.I Aphrodisias still had some access to networks of long-distance trade. One of the pierced shells found in this phase traveled a remarkably long way to arrive at its final deposition spot in Aphrodisias—it is the shell of a Monetaria, a genus of sea snail that originates in the Red Sea (Reese 1986, 192). However, the single shell may have been curated from a previous phase, so it is not, in itself, firm evidence for continued access to trade routes. Better evidence for Aphrodisias's continued access to long-distance trade routes comes from the ceramic assemblage. Although the dining shapes in this phase almost completely belonged to one style, it is also clear that many other ceramic styles were known at Aphrodisias (table 7). Decorative types found in this phase include silver/gold lustrous (Type A1), coppery lustrous (Type A2), linear painted on a lustrous background (Type B1), and white/pale slipped (Type E). What is most interesting here Page 171 →is that while all these styles were present at Aphrodisias, the community here consciously chose not to favor them for use in communal dining activities. As in the previous phase, therefore, contact with external Others did not necessarily mean that the inhabitants of Aphrodisias chose to represent these external Others in their material culture. This seems to have been a conscious choice on the part of the community at Aphrodisias—a decision not to make use of any styles that would have been obviously reminiscent of external Others. Level 5.I Community Identity and Its Social Rationale It may be possible that external relationships were simply not an important source of symbolic currency within the Level 5.I community as they were in earlier phases. If the rough dating of the level to the late tenth century BC is followed, the community of Level 5.I Aphrodisias lived during the most obscure and ill-understood period of the IA, at the time when interregional interaction and economic prosperity in the eastern Mediterranean is thought to have been at its lowest point (see chapter 6). The decreased importance of articulating external links as part of the internal dynamics of the Aphrodisias community may therefore be linked to general decline in contacts and communications hypothesized across Anatolia and the eastern Mediterranean as a whole. As the benefits and frequency of interregional contacts decreased, self-association with the external may have become less socially valuable, and other ways of creating and expressing status may have been sought. The excavated evidence could be taken as suggesting that, like many other communities at this time, Aphrodisias withdrew into itself. Such communities may have become not just more isolated in the practical sense of external contacts and communication but also more socially introverted and no longer outward looking in the way they defined themselves. This would suggest a similar set of social processes happening here as in Level Ia Beycesultan, whereby the population appears not to have been cut off from external contacts but does not appear keen to integrate them symbolically at socially significant occasions. It is equally possible, however, that evidence for community identity may yet be found elsewhere at Aphrodisias, beyond the confines of the small area excavated. It is even possible that external relationships were indeed articulated within the excavated area, but at this relatively late phase of occupation, Page 172 →positioned relatively close to the surface of the Acropolis mound, it may simply be the poor state of archaeological preservation that prevents us from gaining more insight into the Level 5.I community.

Level IIID The Remains of Level IIID The excavators originally divided this phase into five architectural subphases, all of which seem to have followed each other fairly rapidly in time, with no recorded changes in the ceramic repertoire between them. In addition, three of the five original subphases seem to have been somewhat arbitrary stages assigned at the time of excavation, for which Joukowsky could find no architectural distinction when it came to final publication (1986, 125). Moreover, in his original publication of the Trench 6 extension, Marchese described the remains as all belonging to a single archaeological level, Level IIID (1976, 408-10), and Mierse considered the Lydian ceramics from the five subphases as one ceramic assemblage (1986). For the purposes of this case study, I follow Marchese

and Mierse's lead in considering the later IA remains of the Trench 6 extension as a single analytical unit, partly because of the cultural and architectural continuity throughout the subphases, but also, perhaps more important, because of the confusion over whether the initial division into subphases was itself ever valid. Level IIID is separated from Level 5.I by a period of about two centuries, and although there are reports of unstratified sherd material indicating continuity of occupation over this time (Marchese 1986, 109), no architectural remains from these two intervening centuries can be identified. The building in Level IIID was, like its LBA and IA predecessors, made of mudbrick on stone foundations and is thought to have been a fairly large complex with several different rooms (fig. 21). The function of the building remains uncertain, especially as the ceramic assemblage does not suggest the full range of domestic activities, with very few cooking wares found in comparison to the number of dining shapes (table 8). It seems that, like the building excavated in Level 5.I, there might have been more food consumption than food preparation in this area, on a greater scale. It seems probable, therefore, that the Level IIID building was used for the socially charged activity of communal dining and the Page 173 →consumption of food prepared elsewhere on a scale beyond the strictly domestic. However, the building complex seems to have fulfilled a ritual function in addition to being a venue for communal dining. There was an unusual votive deposit sunk into one of the floors, consisting of the skeleton of a dog, an iron dagger, and some associated ceramics (Mierse 1986, 422). This deposit could be interpreted as a one-off foundation deposit, which would have no implications for the function of the building itself. However, the stratigraphy suggests that the deposit was actually dug into the floor of the building. This dog deposit may have been one of several that are yet to be discovered, as it is a form of regular cult deposition known from Sardis (see the following section “Level IIID Representations of the External Other”). This deposit therefore suggests that the building complex was also used for some types of ritual activity, although it is impossible to ascertain what the range of the activities might have been. Page 174 → Overall, then, the structure uncovered in Level IIID seems to have been some sort of central place for the community, where several different types of activity took place, including communal dining and ritual activities. Level IIID Enactments of Community Judging from the absolute numbers of dining vessels uncovered, the feasting events that took place within this structure were on a large scale, involving larger numbers of people than previous events discussed in this book. These events would have been appropriate arenas for the enactment of community identity. However, they would also have been potential arenas for the negotiation of individual status. It is perhaps significant, given this fact, that no items associated with the signaling of personal status (e.g., ornamented weaponry or jewelry) were found (table 9). However, the dining vessels used in the feasting events were not all equal or standardized and therefore did not foster a sense of similarity between the diners. Neither, however, was a diverse set of styles employed in the dining vessels so that each individual or household competed in public display. Instead, save one or two, all the dining vessels from this phase can be divided into two main groups: Type C1 warm slipped wares and Type G3 Lydian wares. This bipartite division suggests that the diners could be divided into two social groups and that the way that feasting events were staged actively worked to promote this division. In Level IIID Aphrodisias, then, feasts were not used to promote either a shared community identity or individual competition for status. Instead, they were used to emphasize a particular division in society that split the diners into two groups. What this division may have meant and what the two groups were can only be inferred by considering how the diners related to external Others and then looking at the wider historical context. Level IIID Representations of the External Other

The most clearly articulated relationship between the inhabitants of Level IIID Aphrodisias and an external Other is obviously with Lydia. The ritual deposit previously mentioned bears striking similarities to contemporary deposits found at Sardis, which have been dubbed “ritual puppy dinners” (Greenewalt 1978). Although the Aphrodisian example is the only such deposit known from Caria, the excavators at Sardis have considered them to be “Carian,” assuming a link with Carian cult figure Page 175 →(Greenewalt 1978, 42; Pedley 1974, 98). Thirty such deposits have been found at Sardis, all of which vary around a standard set of components, including the skeleton of a young canid, four specific ceramic shapes (cooking pot, squat jug, skyphos, and dish), and an iron knife (Greenewalt, 1978, 1). Although the ceramics found with the ritual deposit do not conform to the set prescribed by the Sardian examples, the deposit as a whole is clearly related to those at Sardis. It therefore implies some similarities in cult, even if the precise form of ritual practices were not identical at the two sites (Mierse 1986, 422). This commonality of cult practice is a strong sign of a socially recognized relationship between the Level IIID community at Aphrodisias and the Lydians of the Hermos valley area, a relationship that appears even stronger given the discovery of a Lydian inscription from an unknown location somewhere on the site (Carruba 1970). Although this inscription is later in date than Level IIID, it still indicates that the Lydian script was used at Aphrodisias, and it seems likely that it would have already been in use during the time of the Level IIID occupation. Writing and reading would have been highly socially charged in the IA, when literacy would have been the preserve of an elite group within society. The use of the Lydian script and language would therefore have been a significant social statement. A sense of affinity with Lydia is also articulated in the fineware ceramics of this phase, a good proportion of which are decorated in recognizably Lydian or Lydian-derived styles, Type G3 (Mierse 1986). In addition, the characteristically Lydian vessel shape of the lydion appears, showing the direct adoption of Lydian styles and practices in the socially significant setting of communal dining (Cat. no. A6.167). The remains of Level IIID therefore suggest that a close relationship with Lydia and the area to the immediate northwest was being consciously signaled in this central place and that an awareness of Lydian-influenced visual styles and social practices made a significant contribution to the community's general cultural experience. East Greek visual styles also seem to have had some impact on Level IIID Aphrodisias (table 7). The fineware assemblage includes typical East Greek decorative styles (Type G4), such as the Wild Goat style from Ionia (Cook and Dupont 1998), and Black-on-Red ware, more characteristic of the Greek settlements to the southeast in Cilicia, Cyprus, and northern Syria (Postgate and Thomas 2008, 14; Brodie and Steele 1996). In addition, a number of East Greek vessel shapes appear among the tablewares. These include conical footed cups (Cat. no. A6.140), skyphoi (Cat. nos. Page 176 →A6.83, A6.104, A6.120, A6.157), and Ionian bowls (Cat. nos. A6.28, A6.84). However, it is uncertain whether the inhabitants of Aphrodisias would have associated the Ionian East Greek styles with the Aegean coast itself. They might just as easily have associated them with Lydia, as Ionian styles and shapes are well known at Sardis and seem to have been a fundamental part of the Lydian repertoire (Ramage 1994). Although Aphrodisias could easily have had direct dealings with communities on the coast, it is seems more likely that the Level IIID community would have linked the two conceptually, given the overwhelming strength of the Lydian connection. The decoration of the two Ionian bowls found in Level IIID supports this, as both examples of the characteristically Ionian vessel shape were decorated with a specifically Sardian style of streaky slips. In addition, it has been pointed out that many Lydian and East Greek styles of finewares are closely related and are not always easy to distinguish (DeVries 2000, 356). Besides this clear Lydian connection, Level IIID Aphrodisias also seems to have had connections with other regions, such as Cilicia to the southeast. Although it is classified as an “East Greek” style, Black-on-Red pottery was focused not on the Aegean coast of Anatolia but in southeastern Anatolia, Cyprus, and northern Syria, where the style is associated with Greek colony settlements such as Tarsus and Al-Mina. The Black-on-Red sherd from Level IIID would therefore have evoked associations with the southeast for the community at Aphrodisias and should be seen alongside the Type E white slipped jar with polychrome painted decoration also found in this level (Cat. no. A6.114). These vessels form too small a proportion of the overall assemblage to suggest any meaningful signaling of connections with the southeast, but they are signs that the Level IIID community was in contact with that region and was aware of the stylistic developments there. Similarly, the nine Type D vessels with grey slips

indicate that the community would also maintain an awareness of northwestern Anatolian styles, but there are too few to imply the active articulation of relationships in this direction. A composite style is also evident on the “Carian Jar,” an unusually shaped krater associated with the remains of a cremation found in Trench 6 outside the main area considered in this section (figs. 23a and 22b; Cat. no. A6.171). The fabric of the vessel suggests local production, although a mix of different stylistic influences is evident in its decoration. The main frieze, depicting a bird and lions, is executed in the East Greek Wild Goat style, but decorative filling elements more common to Near Eastern styles, such as the cable motif, are also used. In addition, the style of the lions Page 177 →is said to be “Assyrianizing” (Benjamin 1986, 410-11). The use of a cosmopolitan mixed visual style in the decoration of this vessel, an artifact charged with social significance as a grave marker, is indicative of the outward-looking perspective embraced by the Level IIID community at the turn of the sixth century (Benjamin 1986, 412). The community of Level IIID Aphrodisias therefore seems to have had a wide-ranging set of external relationships and was clearly in contact with a varied set of different external Others. This fits in with the high level of interregional contact and interaction of the time, which saw mercenaries from the general Maeander area involved in the dynastic struggles of Egyptian pharaohs, as well as Assyrian kings corresponding with Lydia (see chapter 6). However, among this broad range of contacts, the inhabitants of Aphrodisias showed a marked preference for Lydian material culture in the socially significant arena of communal dining. The Lydian external Other was favored above the wide range of Level IIID Aphrodisias's other external Others. Lydian styles, however, were not available to all—only about one-third of diners would have used Lydian-style finewares. The remaining two-thirds would mostly have used Type C1 plain slipped wares. Lydian associations were therefore not only favored but also associated with a select group within the population. This group may well have claimed some form of elite or elevated status. This seems likely because of the ornate nature of the Lydian styles and because of their association with the main political power of the time. The embracing of Lydian styles by the Aphrodisias elite and the restriction of their use to only a select group within the population speak volumes about the social dynamics at Level IIID Aphrodisias. Level IIID in Its Historical Context This sense of a close relationship with Lydia and Sardis is to be expected given the historical context. The Lydian state was expanding throughout the period Level IIID was occupied, gradually growing in influence to become the dominant political force in western Anatolia (see chapter 6). Given the proximity of Aphrodisias to the Lydian heartland of the Hermos valley and the long history of connections with this area from the LBA onward, it is therefore not surprising that Aphrodisias was strongly influenced by Lydian styles and social practices in the seventh and sixth centuries. Although the Dandalus valley does not seem to have been part of the central territory of Lydia (Roosevelt 2009, 37), it is very likely that Page 178 →Aphrodisias would have come under the influence of the Lydian state by the mid-sixth century. This influence may have taken the form of direct political control, seeing that the Greek historian Herodotus lists the Carians among those under Lydian suzerainty (1.53.2; see also Balcer 1984), but even if this was not the case, Lydia would have exerted a considerable economic pull on the Dandalus area. Whatever the exact political relationship between Aphrodisias and Sardis was at this time, however, it is likely that the influence of Lydian elite styles would have been felt at Aphrodisias long before any direct control was imposed. The restriction of Lydian styles to the elite at Aphrodisias is surely significant, whether this is to be explained as the result of local elite emulation or of external imposition. It is unfortunate that the finds from Level IIID were not recorded by subphase at the time of excavation, as this may have allowed for a more detailed analysis of the changing ways in which Page 179 →the relationship between Aphrodisias and Lydia was represented at Aphrodisias over this time. However, it does seem clear that the overall picture gained of Level IIID Aphrodisias is a place where clear social distinctions are made. Elites were distinct from non-elites; users of Lydian pottery were marked out from those who used only plain slipped wares.

In some other situations described in this book, it seems that an external threat or the expansion of a powerful neighbor might have encouraged a local community to coalesce around a common community identity. In the case of Level IIID Aphrodisias, however, the opposite seems to have been the case. By making statements in material culture, local elites aligned themselves (or were forcibly aligned) with Lydia, and no community identity or collective sense of togetherness can be discerned.

Conclusions: Communities Broken by Pressure? LBA-IA Aphrodisias As mentioned at the start of this chapter, Aphrodisias is, by its nature, a very different type of site from Beycesultan. Indeed, this chapter has Page 180 →shown that Aphrodisias followed a very different social and historical trajectory. The first LBA-IA phase at Aphrodisias is Level 4.III, which spanned the end of the LBA and the start of the IA. During this time, the community at Aphrodisias banded closely together, engaging in shared communal meals using a highly standardized set of ceramic dining shapes. The strong preference in this phase was for a local style of fineware decoration adapted from a Hermos valley type, while the original Hermos type was also popular. This strong local community identity is perhaps indicative of the unstable times in which the community lived, while the willingness to embrace Hermos valley styles should be seen in the context of the rise of the state of Mira in the Hermos area. During the second LBA-IA phase, Level 4.II, this sense of community identity at Aphrodisias declined. Instead, it seems like there was more emphasis on the articulation of individual identities and status differentiation. This may have been due to the continuing and perhaps strengthening influence of Mira on the site. It certainly seems as if the inhabitants of Aphrodisias had less access to external contacts in this phase than in the previous one and that Aphrodisias's external contacts were then dominated by the Hermos region. Level 4.I, the next phase, sees what might be interpreted as a lifting of this external pressure or influence. For the first time, it appears that Aphrodisias has access to wide-ranging networks of trade and commerce, and the first real signs of prosperity begin to appear. Despite this apparent return to independence, there is no sign of a conscious sense of community identity during this phase. Instead, social differentiation and individual status seem more important than ever. This trend also continues into the next phase, Level 5.II. At this time, the inhabitants of Aphrodisias seem to have continued enjoying their prosperity and access to long-distance trade routes. They also seem to have had little sense of collective group identity, focusing instead on individual status differentiation. During the fifth phase, Level 5.I, this sense of community solidarity may perhaps have returned. Evidence for wealth and external contacts seems to decline slightly, a decline accompanied by a revival of a conscious sense of collective community identity at the site. As this level was occupied at the height of what is known as the Dark Ages, such an inward-looking perspective is perhaps not surprising. As external connections became rarer and political structures became less stable, it is plausible that local communities would have withdrawn into themselves. The final phase, Level IIID, was occupied almost two centuries after the previous phase was abandoned. While scattered sherds suggest that Page 181 →the site may still have been occupied during this time, nothing is known of the intervening gap. The inhabitants of Aphrodisias in Level IIID appear as a complex but not unified population. Social division is stressed between two main groups within society, with the elite members of society perhaps associating themselves with the political power of nearby Lydia, while the majority of the site's population was excluded from this association. The actual political relationship between Aphrodisias and Lydia at this time remains obscure, but it does seem that the community was divided during this final phase and that the Lydians had some part to play in this division. Community Identity in LBA-IA Aphrodisias The preceding summary is at best a narrative based on informed and plausible conjecture. The archaeological remains from Aphrodisias shed light on only a small part of the overall settlement, and conclusions can therefore

only be drawn about some of the practices at Aphrodisias at any given time. However, the practices captured in this small area would have been part and parcel of the wider social dynamics at the site and can therefore tell us something about the social dynamics and trends. Nonetheless, it should still be remembered that what we are seeing at LBA-IA Aphrodisias is far from the complete picture. From this partial picture, however, it does seem that the inhabitants of Aphrodisias, like those of Beycesultan, constructed a sense of conscious community identity at specific times. As at Beycesultan, the practice of feasting or communal dining seems to have been one venue that was sometimes used for staging these enactments of community. However, it appears that community identity became salient at Aphrodisias at slightly different times than it did at Beycesultan. The inhabitants of the two sites reacted differently to the same historical situations, as perhaps might be expected from their different natures and locations. After all, the historical events of the period would have impacted on them in different ways—Beycesultan's prominent location and proximity to the Hittites would ensure that they had a different relationship with central Anatolia from Aphrodisias's; while Aphrodisias's proximity to the Hermos valley was a factor in shaping its tense relationships with Mira and Lydia. For example, the same historical circumstances produced two very different social reactions at Aphrodisias and Beycesultan during the LBA-EIA transition, with people at Aphrodisias coalescing into a community and with those at Beycesultan exploiting new economic and political opportunities at the Page 182 →expense of their neighbors. This can perhaps be explained by the practical differences between the two sites in terms of location and relative importance in the landscape. Beycesultan may have been well placed to capitalize on the opportunities available, while Aphrodisias may simply have been too small and too out of the way to get in on the action. The histories of these two sites are therefore both distinct though related, and each offers an instructive point of comparison for the other. However, the inhabitations of the two sites also seem to have responded in different ways to analogous historical situations. For example, the situation of Beycesultan at the end of the LBA is analogous to that of Aphrodisias at the start of the EIA and in the later IA—the pressure of a Hittite threat over Beycesultan is probably comparable to that of a Miran or Lydian threat over Aphrodisias. But whereas people at Beycesultan seem to have responded to this situation by banding together in the face of a common external enemy, and by fostering a sense of community identity, those at Aphrodisias seem to have reacted quite differently. Instead, status distinctions and hierarchy seem to have become increasingly important at Aphrodisias, with high status often associated with the material culture styles of the external threat itself. Instead of drawing together in the face of a common foe, the inhabitants of Aphrodisias sought to gain status on an individual and subgroup level, using external symbols of power to do this and perhaps currying favor with the external powers in question. This trend at Aphrodisias lies in marked contrast to the analogous response at Beycesultan, where the Hittite threat was met with a deliberate rejection of Hittite styles of material culture. This very different response to an analogous historical context at the two case study sites warns against applying the same rules to all situations. Every community is different, and even two populations living in such close proximity as Aphrodisias and Beycesultan develop a sense of community identity for different reasons and at different times.

Page 183 → CHAPTER NINE

Conclusions Theories Put into Practice The previous two chapters have discussed in detail the LBA and IA remains from the sites of Beycesultan and Aphrodisias, both located inland in western Anatolia (see fig. 1 in chapter 1). These intensive studies have highlighted how material culture was actively used at these sites to construct a conscious sense of community identity. This community identity was neither constant nor natural. It had to be deliberately created and maintained through social practices designed to foster cohesion, stressing similarity and shared experience while downplaying differences within the population. In this book, I have termed these social practices “enactments of community.” These enactments of community involved certain patterns of material culture use, which were traceable in the archaeological record. A sense of collective us was stimulated by bringing together large groups of people (although not necessarily all members of the collective) and by playing up their common experience of shared location and geographical emplacement. At such events, the inhabitants of Aphrodisias and Beycesultan stressed unity and cohesion instead of social differentiation. Participants were treated in a similar fashion, occupying undifferentiated space and making use of standardized and homogeneous forms of material culture. Whatever differences there would have been among segments of the population at the time were glossed over under the rhetoric of community identity. In particular, feasting or communal dining was one activity that was frequently used as an arena for staging enactments of community in both Beycesultan and Aphrodisias. At these events, diners would have eaten their meals in large, undifferentiated spaces and from standardized ceramic vessels. The use of communal service vessels for central, shared dishes of food would also have encouraged group solidarity and the symbolic Page 184 →leveling of social differences. Constructing a sense of community identity is therefore a form of social strategy, an ideology that promotes one particular vision of social relations over others. The studies show that community identity was not consistently pursued as a strategy at Aphrodisias and Beycesultan. Instead, it only became salient in particular social circumstances at particular historical moments. The times when it did crystallize were those times when group solidarity might have had some tangible benefit—when the settlement was under political pressure from a specific external threat at Beycesultan or when the community faced an unstable environment at Aphrodisias. Only at moments such as these did the inhabitants of Aphrodisias and Beycesultan rally around the idea of a collective group identity that brought them together. At other points in time, the inhabitants of these sites were more concerned with the different forms of identity that kept them apart. Rather than community identity, they would choose to emphasize social divisions, such as individual status and gender, age, political allegiances, and especially social rank. The lines by which they differentiated themselves from each other were often more important than the shared experiences that they held in common. Concern with social differentiation was manifested in the way the inhabitants of Aphrodisias and Beycesultan used material culture. During these times, objects that heightened personal status—including such items as ornamented weapons, glass vessels, or imported exotica—became popular and would have created a sense of differential rank. Similarly, greater attention was often paid to the physical body of the individual, and resources were invested in jewelry or other bodily adornments that may have distinguished people along several different lines. Differences in architectural elaboration also acted to entrench, rather than gloss over, social differences. Finally, restricting particular forms of material culture or styles to certain subgroups might have emphasized the distinctiveness of these groups. Exclusive access to certain commodities (e.g., imported prestige items), certain locations (e.g., ritual areas), or certain cultural styles (e.g., types of ceramic decoration) would have strengthened the distinctiveness of certain groups within society. In addition to this sense of collective us, community identity also entailed a sense of external them, or the external Other. At Aphrodisias and Beycesultan, when community identity was crystallized, it was defined in relation to

other communities, and the constructing of an external Other was an important part of rationalizing and understanding the community's own identity. The social rationale of community identity therefore Page 185 →involved defining external Others and the relationships the community had with them. This was done by representing external Others positively or negatively, as closely connected to the community or as distant and estranged. Representations may well have taken verbal or pictorial form, but they could also be performed in social practice, and any archaeological study of community identity must focus on this last form. In particular, objects recalling associations of external Others were deployed meaningfully during enactments of community—at such moments, the collective us was placed in direct contact with the material representation of the external them. Objects or styles that evoked connotations of specific places could be either embraced or ignored, adopted or avoided. Archaeological evidence therefore can be found for a conscious sense of community identity at both Aphrodisias and Beycesultan at different points during the LBA and IA. A sense of community identity is discernible because it necessitated the staging of certain types of social practice, which would have left particular patterns on the material record. These material traces speak eloquently about ideologies of solidarity and distinction, about what is meant by us and what is meant by them. They tell us a great deal about how different social strategies were pursued at the two sites. It is less clear, however, why these different strategies were adopted. Indeed, the case studies warn against there being any simple reason why community identity may crystallize. The two studies show that even at two settlements in the same region, the same historical situations may stimulate different responses depending on the specific nature and location of the site. In addition, it appears that different responses can be expected even in analogous historical situations, where the same pressures are felt. There seems to be no hard and fast rule about when community identity will coalesce or why it might do so.

Community Identity and Our Historical Understanding of Anatolia Aphrodisias and Beycesultan are relatively close to each other geographically and share many overall cultural similarities. Taken together, the evidence from these two sites constitutes a substantial proportion of that currently available for inland western Anatolia during the LBA and IA. This study of community identity at Aphrodisias and Beycesultan can therefore contribute to our historical understanding of LBA-IA Anatolia as a whole. However, it should also be noted that there are many differences Page 186 →between the sites. While Beycesultan is located on a major long-distance route between central and western Anatolia, Aphrodisias is tucked away in a relatively inaccessible valley, leading off south from the main Maeander plain. While Beycesultan was a centralized settlement focused on a single twin-peaked mound, Aphrodisias seems to have been a more dispersed settlement, spread between two, more separate mounds. Differences in excavation history are also important: while large areas of Beycesultan were excavated, recording of finds was selective; at Aphrodisias, a relatively small area was uncovered, but the recording practices were more thorough. The different developments that this book has traced at the two sites are indicative both of the different nature of the available material and of the different nature of the settlements themselves. Our current understanding of LBA-IA western Anatolia is constrained by the idea that the region was caught “between”—between Europe and Asia, East and West, Mycenaeans and Hittites, Greeks and barbarians (see chapter 6). Most existing accounts of this period have taken a top-down approach to western Anatolia, aiming to trace the level of external impact on the region and assuming the primacy of the Aegean and central Anatolia as historical actors. Such approaches reduce western Anatolian populations to the passive receptors of external cultural or political influences, allowing them agency only in the acceptance of or resistance to these influences. Considering community, however, allows us to avoid such assumptions and permits a view of the populations of this region as active and independent. It allows us to see western Anatolian communities not just as either accepting or rejecting outside influences when they are imposed but as actively constructing their own identities from their own social and geographic perspectives. This approach uncovers new insights, not just into the societies of western Anatolia itself, but also into the wider historical situation in Anatolia and the eastern Mediterranean as a whole. A new narrative of the LBA-IA periods can now be told, and a new perspective on Anatolia can now be taken.

A New Narrative for the LBA-IA Most treatments of the LBA period in the region have concluded that western Anatolia was simply beyond the Hittite sphere of direct imperial dominance (e.g., Glatz 2009; Neimeier 1999). My analysis of the community identity expressed at Beycesultan, however, allows us to create a more nuanced picture. Hittite expansionism clearly did have an impact on Beycesultan. Page 187 →The site certainly had contact with the Hittites, as bulk foodstuffs and other commodities were transported between Beycesultan and central Anatolia. However, the impact of the Hittites went far beyond the mere movement of foodstuffs—it had a fundamental effect on the way the inhabitants of Beycesultan related to each other. Crucially, the threat posed by the Hittites seems to have spurred the inhabitants of Beycesultan to band together and to articulate a conscious sense of community identity. There also seems to have been a process of deliberate distancing from the central Anatolian Other: objects with Hittite or central Anatolian connotations were actively rejected, and Hittite styles were studiously ignored. Equally studiously ignored were all associations with areas further west. This is perhaps surprising, as Beycesultan would likely have been allied to these areas as part of the Arzawan confederacy during this time. This rejection of western connections calls into question the strength of the Arzawan confederacy and begs askance of the Hittite king's claims to have crushed a dangerous western alliance. Instead, the community of Beycesultan stressed a specifically local identity, and all outsiders were estranged. The LBA sequence at Aphrodisias opens slightly later than that at Beycesultan, but when it does open, the inhabitants of Aphrodisias also appear to be closely bound by a strong sense of community identity. This community solidarity is perhaps understandable at a time of political instability and transformation. As the LBA moved into the IA, the decline of established kingdoms and the fall of the old political order would have had ripple effects on small and out-of-the-way settlements such as Aphrodisias. It seems that the people of Aphrodisias reacted to the instability by closing ranks. Beycesultan's reaction to the same historical situation was somewhat different, as perhaps might be expected from a larger site on a main trade route. Perhaps taking advantage of the decline of the old powers and the demise of its erstwhile enemy on the Anatolian plateau, Beycesultan prospered over the LBA-IA transition. With the removal of the Hittite threat, Beycesultan lost its sense of community solidarity, and the population instead began to be more concerned with internal distinctions within the community. Rigid social hierarchies and status differentiation took over from community identity. The inhabitants of both sites, however, began to open up their ranks to one particular external Other—the emerging state of Mira in the Hermos valley area. They made enthusiastic use of objects that recalled the Hermos valley, and they adapted Hermos valley styles to suit their own tastes. These developments are a testament to the growth of Mira as a political power in western Anatolia. Page 188 → The growth of Mira seems to have continued into the IA, and Aphrodisias may well have fallen into Mira's expanded political orbit. Community identity at Aphrodisias seems to have been quickly replaced by a concern for social differentiation, and a previously broad network of external contacts now seems to have been dominated by influences from the Hermos valley. This dominance was, however, short lived, and Aphrodisias's access to a wide range of external connections was soon restored. A sense of community identity, however, did not automatically return. As happened at Beycesultan after the lifting of Hittite pressure at the end of the LBA, when Aphrodisias was relieved of pressure from Mira in the IA, its inhabitants became closely concerned with internal differentiation, particularly with distinctions of status or rank. Indeed, an intensified version of this process seems to have been happening at contemporary Beycesultan. During this time, the rigid hierarchies of the initial IA period had given way to a more fluid social structure, where status was acquired competitively. Remains of later IA phases are also found at both sites. At Beycesultan, contacts with Phrygia are evident around the eighth century BC. However, nothing can be concluded either about community spirit or representations of this Phrygian Other, owing to the poor state of preservation of these archaeological levels and the poor standards of recording of the remains. Fortunately, the information available concerning the slightly later seventh-and sixthcentury levels from Aphrodisias is fuller. At this time, Aphrodisias seems to have had a wide range of external contacts, from Cilicia and perhaps even Cyprus in the southeast to the Troad in the northwest. However, of these varied connections, only the Lydian external Other is properly incorporated into Aphrodisias's material repertoire.

The use of Lydian styles seems to have defined a particular group within society, a group that was also made distinct through differential treatment during socially charged activities such as feasting. The Aphrodisian population did not, therefore, maintain a sense of coherent community identity but was, rather, divided into two groups—into the included and the excluded, those using Lydian styles and those not using Lydian styles, the elites and the non-elites. Given the dominant position of Lydia in western Anatolia at the time, Lydian influence was likely to have been felt at Aphrodisias, whether through direct political control or through more indirect forms of cultural authority. In either case, this influence was clearly very strong at Aphrodisias, enough to warrant an elite group defining their difference through the use of Lydian material culture. Page 189 → A New Perspective on the LBA-IA These studies on community identity at Aphrodisias and Beycesultan afford an alternative perspective on LBA-IA western Anatolia. Unlike previous treatments of the region, they do not cast western Anatolia in terms of either Hittites versus Arzawans or Greeks versus barbarians. Instead, they offer a different view of the region—not as being simply “between” but as dynamic and complex in its own right. The nature of the IA is one notable point where this new narrative diverges from the conventional accounts. Traditionally, it was thought that both prosperity and trade declined at the end of the LBA and the start of the IA. However, this view is mostly based on developments seen at the political centers of the LBA. It actually appears that the IA in Anatolia was a time when many settlements, far from falling into poverty and obscurity, took advantage of the collapse of more powerful centers. The inhabitants of sites such as Beycesultan and Aphrodisias may actually have benefited from the fall of nearby powerhouses such as the Hittite Empire and Mira, respectively. The absence of external pressure may not just have allowed the inhabitants of these sites to flourish, however; it may also have allowed them to turn inward. At these moments, a sense of group solidarity and community identity became much less important for the people of both Aphrodisias and Beycesultan, and internal social differentiation came to the fore instead. Another important point of divergence is in the later IA, when conventional narratives imply a strict division between the Hellenic settlements on the Aegean coast and the “native” settlements further inland. However, the material from Aphrodisias suggests that the people here saw little distinction between objects holding Lydian connotations and those with Ionian Greek connotations. The recognized distinction seems to have been not between Lydian styles and Greek styles but, instead, between generic western styles (including both Lydian and Ionian Greek) and southeastern styles (including Levantine Greek and Cypriote). From the perspective of the community at Aphrodisias at least, there seems to have been less distinction between the Greeks and Lydians than many contemporary authors have assumed. Opposition between Ionian Greeks and Lydians is described by Hogarth, for example (1954, 509-20), while Boardman characterizes the treatment of the Ionians by the Lydians as “harsh” (1999, 99). There is a growing body of evidence that argues against this more traditional view, especially as the material excavated at Page 190 →Sardis continues to be studied. In particular, DeVries has remarked that “distinguishing between East Greek and Lydian fine wares is a great challenge to pottery analysts at Anatolian sites” (2000, 356). After all, the image of a strict opposition between Lydians and Greeks has been constructed on the basis of selective interpretations of a small number of selected Greek texts, and there is substantial evidence, even in these texts themselves, of close interaction and cultural similarities between these two groups (Cobet et al. 2007; Greaves 2010; Spawforth 2001). Herodotus, for example, remarks on points of cultural similarity between Lydians and Greeks (1.35), as well as suggesting that Croesus of Lydia was closely involved with the Hellenic Aegean, seeking advice from Solon of Athens (1.30ff.) and dedicating rich gifts at the Panhellenic sanctuary in Delphi (1.51-52). Adopting an approach focusing on community identity has therefore allowed new insights to be gained into western Anatolian protohistory. Instead of viewing it as a region fundamentally characterized by betweenness and instead of focusing on the impact of outside influences on it, looking at community identity permits the region to emerge as a dynamic area in its own right. This approach allows us to move beyond the standard assumptions that

the extent of Hittite influence has to be central to our understanding of the region in the LBA or that the level of Greek influence has to be crucial to characterizing the region in the later IA. Instead, it places the societies of western Anatolia on center stage.

Community Identity and Archaeology The study of community identity in archaeology has great potential to give us a more nuanced understanding of ill-understood sites and to offer alternative perspectives on already-familiar histories. In the case of the two sites considered in this book, Aphrodisias and Beycesultan, considering community identity has fulfilled both aspects of this potential. Insights have been gained both into the internal social dynamics at the sites themselves and into the wider historical processes that affected the rest of Anatolia and the eastern Mediterranean as a whole. Such insights can only be gained, however, if community identity is theorized carefully and investigated following a rigorous methodology. In this book, I have suggested both a theoretical and methodological framework for approaching community identity through material culture. Initially, I considered the concept of the community, emphasizing Page 191 →the distinction between relational communities, which include all groups of shared interest, and geographic communities, which are a specific form of relational community involving a shared geographic focus and coresidence. As “relational community” is an unhelpfully vague term, applicable to any form of generic group identity, it is explicitly stated that geographic communities are the focus for this particular book. Community identity should be understood as a socially constructed ideology, whereby inhabitants of a site are encouraged to feel a sense of group solidarity in opposition to an external Other. This sense of collective us is partly based on the common experiences of coresidence and shared emplacement in the wider landscape. However, a sense of group solidarity does not arise naturally from these experiences—it is merely facilitated by them. Group solidarity must be actively constructed through social practice, and a sense of community identity must be crystallized through enactments of community. In these enactments, internal distinctions within the community must be glossed over, and differences must be played down. Members of the community must be made to feel the primacy of their shared community membership over and above the various differences they may have from each other. However, fostering this sense of us is only one part of community identity. The idea of us must also be constructed in relation to an idea of them. The difference between us and them is rationalized through a situationspecific social rationale. This social rationale will recognize the local nature of community identity but may also posit a variety of secondary features of this community identity—for example, kinship and ethnicity, socioeconomic status or professional occupation, religious practice or political affiliation. The definition of “them” along the lines of this social rationale is a central part of determining who the us of community identity really is. Identifying traces of community identity in the archaeological record is as big a challenge as theorizing it. To infer an active sense of community identity at a site, it is necessary first to find evidence for enactments of community. These can take the form of traces of large gatherings being held in open spaces. These spaces would need to accommodate a reasonably sized group, without separating them or encouraging differentiation. The objects used in enactments of community would also need to discourage a sense of difference and so would probably be standardized in type and form. Beyond these enactments of community, there may also be other circumstantial evidence for community identity at a given site. The inhabitants at the site may show less sign of emphasizing social distinctions; Page 192 →there may be relatively few cases of displaying individual status through personal accessories, for example, or of flaunting household wealth through architectural elaboration. If there is evidence for a conscious sense of community identity at a site, it is important to consider how this community constructed a sense of us in relation to an external them. The social rationale of the community identity can be investigated by analyzing the way external Others are represented within the community—which Others are acknowledged during enactments of community in particular and which are not. How the community

thought about outsiders and which outsiders they chose to associate themselves with tells us much about how they rationalized their own community identity. This can be seen in the way that objects were used during enactments of community that recall associations of other places. It is evident that some objects can evoke a sense of a particular external Other either by their form or style, and the way that these objects were deployed in socially charged events is therefore significant of how these particular external Others were viewed. It is therefore possible to use archaeological material to consider whether the inhabitants of a given site had a conscious sense of community identity and what this community identity may have meant. However, the archaeological remains of the site in question may not, on their own, explain why community identity may have crystallized at a particular time or why it may have declined. For this, some knowledge of the wider historical context is necessary, to situate the specific social dynamics of the site within the wider dynamics of the region. In this book, I have attempted to put the study of community identity back on the archaeological agenda. Previous attempts to do this have resulted in pioneering works of crucial importance, but these have often been met with criticism and skepticism. However, this book asserts that concerns over the feasibility or theoretical robustness of such studies can be addressed if the concept of community identity is approached in a carefully defined way. If properly conceptualized, community identity can indeed offer an exciting new approach to past societies, presenting unforeseen perspectives on historical events and uncovering new insights into social dynamics and strategies. The two sites considered here are clear illustrations of the great potential that the study of community identity has for archaeology: our understanding has been increased concerning the developments at two important but hitherto ill-understood sites, and conventional narratives about the eastern Mediterranean in the LBA and IA have been challenged. Page 193 → However, the archaeological study of community identity is important not just because it offers us new insights into the societies of the past. As discussed in chapter 1 of this book, the idea of the local community is becoming increasingly important to us in the changing world of the twenty-first century. By looking at the communities of the past, we can hope to gain a better understanding of communities in the present and perhaps also in the future.

Page 194 → Page 195 → Page 196 → Page 197 → Page 198 → Page 199 →

Page 200 → Page 201 →

Appendixes The appendixes to this book present the first comprehensive catalog of all known small finds and ceramics from the LBA and EIA levels of both Aphrodisias and Beycesultan. They bring together published data as well as information from archival records and primary analysis. See the chapters on the individual sites for more details. In this catalog, each individual object is assigned a catalog number distinct from its original excavation reference. This has been done to allow for standardization and simplicity in the referencing of objects in this book. Similarly, the different phases at each site have been allocated shorthand references for use in the catalog, although they will be referred to in the main text by the terms designated by the excavators rather than by this shorthand. The phases of the sites are designated in the catalog as follows (earliest to latest): Beycesultan Level III

B1

Level II

B2

Level Ib

B3

Level Ia

B4

Phrygian Level

B5

Aphrodisias Level 4.III

A1

Level 4.II

A2

Level 4.I

A3

Level 5.II

A4

Page 202 → Level 5.I

A5

Level IIID

A6

Catalog numbers for ceramic vessels consist of the level in which the vessel was found followed by an Arabic numeral after a period (e.g., A4.61). Catalog numbers for small finds consist of the level in which the object was found followed by an Arabic numeral after a hyphen (e.g., A4-61).

Page 203 → APPENDIX A Small Finds from Beycesultan Page 204 → Page 205 → Page 206 → Page 207 → Page 208 →

Page 209 → APPENDIX B Ceramics from Beycesultan Page 210 → Page 211 → Page 212 → Page 213 → Page 214 → Page 215 → Page 216 → Page 217 → Page 218 → Page 219 → Page 220 → Page 221 →

Page 222 → APPENDIX C Small Finds from Aphrodisias Page 223 → Page 224 → Page 225 → Page 226 → Page 227 → Page 228 → Page 229 → Page 230 →

Page 231 → APPENDIX D Ceramics from Aphrodisias Page 232 → Page 233 → Page 234 → Page 235 → Page 236 → Page 237 → Page 238 → Page 239 → Page 240 →

Page 241 →

Notes CHAPTER

1

1. An archaeology of communities is explicitly different from community archaeology. The former is concerned with the archaeological investigation of historic and prehistoric communities, whereas the latter is concerned with engaging the public in archaeology, particularly with encouraging local people to take an active interest in their heritage. CHAPTER

7

1. Excavation dates are given in Seton Lloyd's personal diaries as follows: 1954: 05/05–28/06 (fifty-four days); 1955: 01/05–06/07 (sixty-eight days); 1956: 02/05–13/07 (forty-one days); 1957: 07/05–06/07 (sixty days); 1958: 23/05–20/06 (twenty-eight days); 1959: 15/09–28/10 (forty-four days). 2. Access to the material in the museum was kindly granted by the Turkish Ministry of Culture, and facilitated by Dr Hüseyn Baysal, the museum director, and Salim U ulmar, the assistant keeper. CHAPTER

8

1. Access to the on-site storerooms at Aphrodisias was kindly granted by Christopher Ratté and facilitated by Bert Smith, who, along with the rest of the Aphrodisias team, were very helpful in solving all my practical problems on-site.

Page 242 → Page 243 →

Bibliography Abay, E., and F. Dedeo lu. 2005. “2003 Yili Denizli/Çivril Ovasi Yüzey Ara tirmasi.” Kazi Sonuçlari Toplantisi 22:41–50. Abay, E., and F. Dedeo lu. 2007. “2005 Yili Çivril Ovasi Yüzey Ara tirmasi 2005.” Kazi Sonuçlari Toplantisi 24:277–92. Abay, E., and F. Dedeo lu. 2009. “Beycesultan 2007–2008 Yilari Kazi Çali malari Ön Raporu.” Arkeologi Dergisi 8:53–80. Abu-Lughod, J. L., ed. 1994. From Urban Village to East Village. Oxford: Blackwell. Adiego Lajara, I–J. 2007. The Carian Language. Leiden: Brill. Akurgal, E. 1962. “The Early Period and the Golden Age of Ionia.” American Journal of Archaeology 66:369–79. Alp, S. 1968. Zylinder und stempelsiegel as Karaköyük bei Konya. Ankara: Türk Tarih Kurumu. Amit, V., ed. 2002. Realizing Community: Concepts, Social Relationships, and Sentiments. London: Routledge. Amit, V., and N. Rapport. 2002. The Trouble with Community: Anthropological Reflections on Movement, Identity, and Collectivity. London: Pluto. Anderson, B. 1983. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. Appadurai, A. 1988. “Introduction: Commodities and the Politics of Value.” In A. Appadurai, ed., The Social Life of Things: Commodities in Cultural Perspective, 3–65. Cambridge: Cambridge University Press. Appadurai, A. 1996. Modernity at Large: Cultural Dimensions of Globalization. Minneapolis: University of Minnesota Press. Artzy, M. 2001. “White Slip Ware for Export: The Economics of Production.” In V. Karageorghis, ed., The White Slip Ware of Late Bronze Age Cyprus, 107–15. Vienna: Verlag der Österreichschen Akademie der Wissenschaften. Bachhuber, C., and R. G. Roberts, eds. 2009. Forces of Transformation: The End of the Bronze Age in the Mediterranean. Oxford: Oxbow. Bahrani, Z. 1998. “Conjuring Mesopotamia: Imaginative Geography and a World Past.” In L. Meskell, ed., Archaeology under Fire: Nationalism, Politics,Page 244 → and Heritage in the Eastern Mediterranean and Middle East, 159–74. London: Routledge. Bakir, T. 1997. “Phryger in Daskyleion.” In R. Gusmani, M. Salvini, and P. Vannicelli, eds., Frigi e Frigio, 229–38. Rome: Consiglio Nazionale delle Ricerche. Bal, M., and N. Bryson. 1991. “Semiotics and Art History.” The AA Bulletin 73:174–208. Balcer, J. M. 1984. Sparda by the Bitter Sea: Imperial Interaction in Western Anatolia. Chico, CA: Scholars Press. Barnett, R. D. 1975a. “Phrygia and the Peoples of Anatolia in the Iron Age.” In I. E. S. Edwards, N. G. L. Hammond, and E. Sollberger, eds., The Cambridge Ancient History, vol. 2, part 2, The Middle East and the Aegean Region, c.1380–1000 B.C., 417–42. Cambridge: Cambridge University Press.

Barnett, R. D. 1975b. “The Sea Peoples.” In I. E. S. Edwards, N. G. L. Hammond, and E. Sollberger, eds., The Cambridge Ancient History, vol. 2, part 2, The Middle East and the Aegean Region, c.1380–1000 B.C., 359–78. Cambridge: Cambridge University Press. Barth, F., ed. 1969. Ethnic Groups and Boundaries: The Social Organization of Culture Difference. Boston: Little, Brown. Basedow, M. A. 2000. Be ik-Tepe: Das spätbronzezeitliche Gräberfeld. Mainz: Verlag Phillipp von Zabern. Basedow, M. A. 2002. “Cemetery and Ideology in the West Anatolian Coastal Region.” In R. Aslan, S. Blum, G. Kastl, F. Schweizer, and D. Thumm, eds., Mauerschau: Festschrift für Manfred Korfman, 469–74. RemshaldenGrunbach: Verlag Bernhard Albert Greiner. Bass, G. F. 1987. “Oldest Known Shipwreck Reveals Splendours of the Bronze Age.” National Geographic Magazine 172, no. 6 (December): 692–733. Bass, G. F. 1991. “Evidence of Trade from Bronze Age Shipwrecks.” In N. H. Gale, ed., Bronze Age Trade in the Mediterranean, 59–82. Jonsered: Paul Åströms Förlag. Bayne, N. P. 2000. The Grey Wares of North-West Anatolia in the Middle and Late Bronze Age and Early Iron Age and Their Relation to the Early Greek Settlements. Asia Minor Studien. Bonn: Dr. Rudolph Habelt GMBH. Beckman, G. 1996. Hittite Diplomatic Texts. Atlanta, GA: Scholars Press. Becks, R. 2003. “Troia VII: The Transition from the Late Bronze Age to the Early Iron Age.” In B. Fischer, H. Genz, É. Jean, and K. Köro lu, eds., Identifying Changes: The Transition from Bronze to Iron Ages in Anatolia and the Neighbouring Regions, 41–53. Istanbul: Ege Yayinlari. Bell, C. 2009. “Continuity and Change: The Divergent Destinies of Late Bronze Age Ports in Syria and Lebanon across the LBA/Iron Age Transition.” In C. Bachhuber and R. G. Roberts, eds., Forces of Transformation: The End of the Bronze Age in the Mediterranean, 30–38. Oxford: Oxbow. Bell, C., and H. Newby. 1971. Community Studies. London: Unwin. Bell, C., and H. Newby, eds. 1974. The Sociology of Community. London: Routledge. Bell, D., and G. Valentine. 1997. Consuming Geographies: We Are Where We Eat. London: Routledge. Page 245 → Benjamin, A. S. 1986. “The ‘Carian’ Jar.” In M. S. Jouskowsky, Prehistoric Aphrodisias: An Account of the Excavations and Artifact Studies, 408–12. Louvain-la-Neuve: Institut supérieur d'archéologie et d'histoire de l'art, Collège Érasme. Bennet, J., and J. Davies. 1999. “Making Mycenaeans: Warfare, Territorial Expansion, and Representation of the Other in the Pylian Kingdom.” In R. Laffineur, ed., Polemos: Le contexte guerrier en Égée à l'Âge du Bronze, 105–20. Aegaeum 19. Liège: Université de Liège. Benzi, M. 2002. “Anatolia and the Eastern Aegean at the Time of the Trojan War.” In F. Montanari, ed., Omero Tremila Anni Dopo, 343–404. Storia e Letteratura: Raccolta di Studi e Testi 210. Rome: Edizioni Storia e Letteratura. Bhabha, H. K. 1990. “DissemiNation: Time, Narrative, and the Margins of the Modern Nation.” In H. K. Bhabha, ed., Nation and Narration, 291–322. London: Routledge. Bhabha, H. K. 1994. The Location of Culture. London: Routledge.

Binford, L. R. 1962. “Archaeology as Anthropology.” American Antiquity 28:217–25. Binford, L. R. 1968. “Archaeological Perspectives.” In L. R. Binford and S. R. Binford, eds., New Perspectives in Archaeology, 5–32. Somerset, NJ: Aldine Transaction. Blackman, M. J. 1986. “The Provenance of Obsidian Artefacts from Late Chalcolithic Levels at Aphrodisias.” In M. S. Joukowsky, Prehistoric Aphrodisias: An Account of the Excavations and Artifact Studies, 279–85. Louvainla-Neuve: Institut supérieur d'archéologie et d'histoire de l'art, Collège Érasme. Blegen, C. W., C. G. Boulter, J. L. Caskey, and M. Rawson. 1958. Troy, Volume 4: Settlements VIIa, VIIb, and VIII. Princeton, NJ: Princeton University Press. Blegen, C. W., J. L. Caskey, and M. Rawson. 1953. Troy, Volume 3: The Sixth Settlement. Princeton, NJ: Princeton University Press. Bliege Bird, R., and E. Alden Smith. 2005. “Signalling Theory, Strategic Interaction, and Symbolic Capital.” Current Anthropology 46:221–48. Boardman, J. 1964. The Greek Overseas. 1st ed. Harmondsworth, Middlesex: Penguin Books. Boardman, J. 1999. The Greeks Overseas. 4th ed. London: Thames and Hudson. Bourdieu, P. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press. Briant, P. 1996. Histoire de L'Empire Perse de Cyrus à Alexandre. Paris: Fayard. Brixhe, C. 2004. “Phrygian.” In R. D. Woodward, ed., The Cambridge Encyclopedia of the World's Ancient Languages, 777–88. Cambridge: Cambridge University Press. Brixhe, C., and M. Lejeune. 1984. Corpus des inscriptions paléo-phrygiennes. Paris: Institut français d'études anatoliennes d'Istanbul. Brodie, N. J., and L. Steele. 1996. “Cypriot Black-on-Red Ware: Towards a Characterization.” Archaeometry 38:263–78. Brodsky, A. 1996. “Resilient Single Mothers in Risky Neighbourhoods: Negative Psychological Sense of Community.” Journal of Community Psychology 24:347–63. Page 246 → Brody, H. 1981. Maps and Dreams: Indians and the British Columbia Frontier. Vancouver: Douglas and MacIntyre. Broodbank, C. 2004. “Minoanisation.” The Proceedings of the Cambridge Philological Society 50:46–91. Bruhn, J. G. 2005. The Sociology of Community Connections. Berlin: Springer. Bryce, T. 2006. Trojans and Their Neighbours. London: Routledge. Buber, M. 1947. Between Man and Man. Translated by R. G. Smith. London: Routledge and Kegan Paul. Buber, M. 1958. I and Thou. Translated by R. G. Smith. New York: Charles Scribner's Sons. Bulmer, M. 1984. The Chicago School of Sociology: Institutionalization, Diversity, and the Rise of Sociological Research. Chicago: University of Chicago Press. Büyükkolanci, M. 2000. “Excavations on Ayasuluk Hill in Selçuk/Turkey: A Contribution to the Early History of

Ephesus.” In F. Krinzinger, ed., Die Ägäis und das westliche Mittelmeer: Beziehungen und Wechselwirkungen 8. bis 5. Jh. v. Chr., 39–43. Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Calame, C. 1990. “Narrating the Foundation of a City: The Symbolic Birth of Cyrene.” In L. Edmunds, ed., Approaches to Greek Myth, 277–341. Baltimore: Johns Hopkins University Press. Cameron, A., ed. 1989. History as Text: The Writing of Ancient History. London: Duckworth. Canuto, M. A., and J. Yaeger, eds. 2000. The Archaeology of Communities: A New World Perspective. London: Routledge. Carruba, O. 1970. “A Lydian Inscription from Aphrodisias in Caria.” Journal of Hellenic Studies 90:195–96. Chabot Aslan, C. 2009. “End or Beginning? The Late Bronze Age to Iron Age Transformation at Troia.” In C. Bachhuber and R. G. Roberts, eds., Forces of Transformation: The End of the Bronze Age in the Mediterranean, 142–49. Oxford: Oxbow. Chang, K. C. 1968. Introduction to K.C. Chang, ed., Settlement Archaeology, 1–19. Palo Alto, CA: National Press. Chavis, D., J. Hogge, D. McMillan, and A. Wandersman. 1986. “Sense of Community through Brunswick's Lens: A First Look.” Journal of Community Psychology 14:24–40. Childe, G. V. 1929. The Danube in Prehistory. Oxford: Oxford University Press. Childe, G. V. 1958. The Prehistory of European Society. London: Pelican Books. Chorney, H. 1990. City of Dreams: Social Theory and the Urban Experience. Nelson, Canada: Scarborough. Cline, E. H. 1994. Sailing the Wine-Dark Sea. BAR International Series, 591. Archaeopress. Cline, E. H. 2008. “Troy as a ‘Contested Periphery’: Archaeological Perspectives on Cross-Cultural and CrossDisciplinary Interactions Concerning Bronze Age Anatolia.” In B. J. Collins, M. R. Bachvarova, and I. C. Rutherford, eds., Anatolian Interfaces: Hittites, Greeks, and Their Neighbours, 11–20. Oxford: Oxbow. Page 247 → Cline, E. H., and H. Harris-Cline, eds. 1998. The Aegean and the Orient in the Second Millennium. Aegaeum 18. Liège: Université de Liège. Cobet, J., V. von Graeve, W.-D. Niemeier, and K. Zimmerman, eds. 2007. Frühes Ionien Eine Bestandsaufnahme. Akten des Symposions: 100 Jahre Milet, 1999. Mainz: Philip von Zabern. Cogan, M., and H. Tadmor. 1977. “Gyges and Ashurbanipal: A Study in Literary Transmission.” Orientalia 46:65–85. Cohen, A. P. 1982. Belonging: Identity and Social Organisation in British Culture. Manchester: Manchester University Press. Cohen, A. P. 1985. The Symbolic Construction of Community. London: Routledge. Cohen, A. P. 1993. Masquerade Politics: Explorations in the Structure of Urban Cultural Movements. Berkeley: University of California Press. Cohen, A. P. 2002. “Epilogue.” In V. Amit, ed., Realizing Community: Concepts, Social Relationships, and Sentiments, 165–70. London: Routledge.

Cohen, B. 2000. Introduction to B. Cohen, ed., Not the Classical Ideal: Athens and the Construction of the Other in Greek Art, 1–20. Leiden: Brill. Coldstream, J. N. 1968. Greek Geometric Pottery. London: Methuen. Coldstream, J. N. [1977] 2003. Geometric Greece, 900–700 B.C. 2nd ed. London: Routledge. Cook, J. M. 1958–59. “Old Smyrna, 1948–1951.” Annual of the British School at Athens 53–54:1–34. Cook, J. M. 1962. The Greeks in Ionia and the East. London: Thames and Hudson. Cook, J. M. 1975. “Greek Settlement in the Eastern Aegean and Asia Minor.” In I. E. S. Edwards, N. G. L. Hammond, and E. Sollberger, eds., The Cambridge Ancient History, vol. 2, part 2, The Middle East and the Aegean Region, c.1380–1000 B.C., 773–804. Cambridge: Cambridge University Press. Cook, R. M., and P. Dupont. 1998. East Greek Pottery. London: Routledge. Crielaard, J.-P. 2009. “The Ionians in the Archaic Period: Shifting Identities in a Changing World.” In T. Derks and N. Roymans, eds., Ethnic Constructs in Antiquity: The Role of Power and Tradition, 37–84. Amsterdam: Amsterdam University Press. Danahay, M. A. 1993. A Community of One: Masculine Autobiography and Autonomy in Nineteenth-Century Britain. New York: State University of New York Press. Davies, G. 1994. A History of Money: From Ancient Times to the Present Day. Cardiff: University of Wales Press. Davies, W. V., and L. Schofield, eds. 1995. Egypt, the Aegean, and the Levant: Interconnections in the Second Millennium BC. London: British Museum Press. Dawson, A. 2002. “The Mining Community and the Ageing Body: Towards a Phenomenology of Community?” In V. Amit, ed., Realizing Community: Concepts, Social Relationships, and Sentiments, 21–37. London: Routledge. De Boech, F. 1998. “The Rootedness of Trees: Place as Cultural and Natural Texture in Rural Southwest Congo.” In N. Lovell, ed., Locality and Belonging, 25–54. London: Routledge. Page 248 → Deetz, J. 1977. “Material Culture and Archaeology: What's the Difference?” In L. Fergusson, ed., Historical Archaeology and the Importance of Material Things, 9–12. Chelsea, NY: Society for Historical Archaeology. Delanty, G. 2003. Community. London: Routledge. DeMarrais, E. 2002. “Beyond the Individual.” Cambridge Archaeological Journal 12:69–71. DeMarrais, E., L. J. Castillo, and T. Earle. 1996. “Ideology, Materialization, and Power Strategies.” Current Anthropology 37:15–31. Desborough, V. R. D. A. 1972. The Greek Dark Ages. London: Benn. DeVries, K. 2000. “The Nearly Other: The Attic Vision of Phrygians and Lydians.” In B. Cohen, ed., Not the Classical Ideal: Athens and the Construction of the Other in Greek Art, 338–63. Leiden: Brill. Dewdney, J. C. 1971. Turkey. London: Chatto and Windus. Diaz-Andreu, M., S. Lucy, S. Babic, and D. N. Edwards. 2005. The Archaeology of Identity: Approaches to Gender, Age, Status, Ethnicity, and Religion. London: Routledge.

Dickinson, O. T. P. K. 1994. The Aegean Bronze Age. Cambridge: Cambridge University Press. Dietler, M. 1990. “Driven by Drink: The Role of Drinking in the Political Economy and the Case of Early Iron Age France.” Journal of Anthropological Archaeology 9:352–406. Dietler, M. 1999. “Rituals of Commensality and the Politics of State Formation in the ‘Princely’ Societies of Early Iron Age Europe.” In P. Ruby, ed., Les princes de la Protohistoire et l'émergence de l'état, 135–52. Naples: Collection de l'École Française de Rome. Dietz, U. L. 1999. “Horses in the Bronze Age.” In K. Demakopoulou, C. Eluère, J. Jensen, A. Jockenhövel, and J.-P. Mohen, eds., Gods and Heroes of the European Bronze Age, 83–84. London: Thames and Hudson. Dobres, M. A., and J. E. Robb. 2000. Agency in Archaeology. London: Routledge. Dornan, J. L. 2002. “Agency and Archaeology: Past, Present, and Future Directions.” Journal of Archaeological Method and Theory 9:303–29. Dougherty, C. 1993. The Poetics of Colonization: From City to Text in Archaic Greece. Oxford: Oxford University Press. Düring, B. S. 2006. Constructing Communities: Clustered Neighborhood Settlements of the Central Anatolian Neolithic. Leiden: Nederlands Instituut vor het Nabije Oosten. Durkheim, É. 1893. De la division du travail social: Étude sur l'organisation des sociétés supérieures.Paris: Alcan. Durkheim, É. 1912. Les formes élémentaires de la vie religieuse: Le système totémique en Australie.Paris: Les Presses universitaires de France. Dyson, S. L. 1992. Community and Society in Roman Italy. Baltimore: John Hopkins University Press. Dyson, S. L. 2006. In Pursuit of Ancient Pasts: A History of Classical Archaeology in the Nineteenth and Twentieth Centuries. New Haven: Yale University Press. Easton, D., J. D. Hawkins, A. Sherratt, and S. Sherratt. 2002. “Troy in Recent Perspective.” Anatolian Studies 52:75–109. Page 249 → Ebbinghaus, S. 2008. “Patterns of Elite Interaction: Animal-Headed Vessels in Anatolia in the Eighth and Seventh Centuries BC.” In B. J. Collins, M. R. Bachvarova, and I. C. Rutherford, eds., Anatolian Interfaces: Hittites, Greeks, and Their Neighbours, 181–90. Oxford: Oxbow. Ebeling, E., B. Meissner, E. Weidner, W. von Soden, and D. O. Edzard, eds. 1993. Reallexikon der Assyriologie und Vorderasiatischen Archäologie. Vol. 8. Berlin: Walter de Gruyter. Emlyn-Jones, C. J. 1980. The Ionians and Hellenism. London: Routledge and Kegan Paul. Eriksson, K. 1993. Red Lustrous Wheel-Made Ware. Jonsered: Paul Åströms Förlag. Erim, K. T. 1976. “De Aphrodisiade.” American Journal of Archaeology 71:233–43. Erkanal, A., and H. Erkanal. 1986. “A New Archaeological Excavation in Western Anatolia: Panaztepe.” Turkish Review Quarterly Digest, Spring, 67–78. Ersoy, Y. E. 2004. “Klazomenai, 900–500 BC: History and Settlement Evidence.” In A. Moustaka, E. Skarlatidou, M.-C. Tzannes, and Y. E. Ersoy, eds., Klazomenai, Teos, and Abdera: Metropolis and Colony, 43–75.

Thessaloniki: University Studio Press. Etzioni, A. 1993. The Spirit of the Community. New York: Crown. Evans, J. 2004. Review of Orhan Pamuk, Snow. New Statesman (London), May 10. Finkelberg, M. 1988. “From Ahhiyawa to Achaioi.” Glotta 66:127–34. Finley, M. I. 1985. “The Ancient Greeks and Their Nation.” In M. I. Finley, ed., The Use and Abuse of History, 120–33. London: Chatto and Windus. Fischer, F. 1963. Die Hethitishe Keramik von Bo azköy. Berlin: Wissenschaftliche Veröffentlichungen der Deutschen Orient-Gesellschaft. Flannery, K., ed. 1976. The Early Mesoamerican Village. New York: Academic Press. Fowler, C. 2004. The Archaeology of Personhood: An Anthropological Approach. London and New York. Frankenburg, R. 2004. “British Community Studies: Problems of Synthesis.” In M. Banton, ed., The Social Anthropology of Complex Societies, 123–53. London: Routledge. French, D. 1981. Roman Roads and Milestones in Asia Minor. BAR International Series, 392. Oxford: Archaeopress. Gale, N. H., ed. 1991. Bronze Age Trade in the Mediterranean. Jonsered: Paul Åströms Förlag. Galpin, C. J. 1915. The Social Anatomy of an Agricultural Community. Madison: University of Wisconsin. Gamble, C. 2001. Archaeology: The Basics. London: Routledge. Gamble, C. 2007. Origins and Revolutions: Human Identity in Earliest Prehistory. Cambridge: Cambridge University Press. Gardner, A. 2007. An Archaeology of Identity: Soldiers and Society in Later Roman Britain. Walnut Creek, CA: Left Coast Press. Garstang, J., and O. Gurney. 1959. The Geography of the Hittite Empire. London: British Institute of Archaeology at Ankara. Page 250 → Gates, C. 1995. “Defining Boundaries of a State: The Mycenaeans and Their Anatolian Frontier.” In R. Laffineur and W.-D. Niemeier, eds., Politeia: Society and State in the Aegean Bronze Age, 289–98. Aegaeum 12. Liège: Université de Liège. Gates, M.-H. 2001. “Potmarks at Kinet Hüyük and the Hittite Ceramic Industry.” Varia Anatolica 13:137–57. Geogiadis, M. 2003. The South-Eastern Aegean in the Mycenaean Period: Islands, Landscape, Death, and Ancestors. BAR International Series 1196. Oxford: Archaeopress. Gerritsen, F. 2004. “Archaeological Perspectives on Local Communities.” In J. L. Bintliff, ed., A Companion to Archaeology, 141–54. Oxford: Blackwell. Gibson, J. J. 1977. “The Theory of Affordances.” In R. Shaw and J. Bransford, eds., Perceiving, Acting, and Knowing: Toward an Ecological Psychology, 67–82. Hillsdale, NJ: Erlbaum.

Giddens, A. 1984. The Constitution of Society: Outline of the Theory of Structuration. Cambridge: Polity. Gitin, S. 1995. Recent Excavations in Israel: A View to the West. Dubuque, IA: Kendall Hunt. Givanni, J. 1996. “Horace Ové: Reflection on a Thirty-Year Experience.” Black Film Bulletin, Summer, 16–21. Glatz, C. 2006. “Contact, Interaction, Control: The Archaeology of Inter-regional Relations in Late Bronze Age Anatolia.” PhD diss., Institute of Archaeology, University College London. Glatz, C. 2009. “Empire As Network: Spheres of Material Interaction in Late Bronze Age Anatolia.” Journal of Anthropological Archaeology 28:127–41. Glynn, T. 1986. “Neighbourhood and Sense of Community.” Journal of Community Psychology 14:341–52. Goldstein, P. S. 2000. “Communities without Borders: The Vertical Archipelago and Diaspora Communities in the Southern Andes.” In M. A. Canuto and J. Yaeger, eds., The Archaeology of Communities: A New World Perspective, 182–209. London: Routledge. Gosden, C. 1994. Social Being and Time. Oxford: Blackwell. Gray, J. 2002. “Community as Place-Making: Ram Auctions in the Scottish Borderland.” In V. Amit, ed., Realizing Community: Concepts, Social Relationships, and Sentiments, 38–59. London: Routledge. Greaves, A. M. 2002. Miletos: A History. London: Routledge. Greaves, A. M. 2007. “Trans-Anatolia: Examining Turkey as a Bridge between East and West.” In A. Irving and A. Greaves, eds., Transanatolia: Examining Turkey as a Bridge between East and West, 1–15. Anatolian Studies 57. London: British Institute of Archeaology at Ankara. Greaves, A. M. 2009. “Bronze Age Aphrodisias Revisited.” In C. Gallou, M. Georgiadis, and G. M. Muskett, eds., Dioskouroi. Studies Presented to W. G. Cavanagh and C. B. Mee on the Anniversary of Their 30-Year Joint Contribution to Aegean Archaeology. BAR International Series, 1889. Oxford: Archaeopress. Greaves, A. M. 2010. The Land of Ionia: Society and Economy in the Archaic Period. Oxford: Blackwell. Page 251 → Greene, K. 2002. Archaeology: An Introduction. 4th ed. London: Routledge. Greenewalt, C. H. 1966. “Lydian Pottery of the Sixth Century B.C.: The Lydion and Marbled Ware.” PhD diss., University of Pennsylvania. Greenewalt, C. H. 1968. “Lydian Vases from Western Asia Minor.” California Studies in Classical Antiquity 1:139–54. Greenewalt, C. H. 1978. Ritual Dinners in Early Historic Sardis. Los Angeles: University of California Press. Grélois, J.-P. 1988. “Les Annales décennales de Mursili II.” Hethitica 9:17–146. Griffiths Pedley, J. 1968. Sardis in the Age of Croesus. Norman: University of Oklahoma Press. Günel, S. 1999. Panaztepe II: M.Ö. 2. Bine Tarihlendirilen Panaztepe Seramiginin Bati Anadolu ve Ege Arkeolojisindeki Yeri ve Önemi / Die Keramik von Panaztepe und ihre Bedeutung für Westkleinasien und die Ägäis im 2. Jahrtausend. Ankara: Türk Tarih Kurumu. Günel, S., and S. Herhordt. 2010. “Ein hethitischer Siegealabdruck aus Çine-Tepecik.” Archäologischer Anzeiger 2010/1:1–11.

Gunter, A. 2006. “Issues in Hittite Ceramic Production: A View from the Western Frontier.” In D. P. Mielke, U.-D. Schoop, and J. Seeher, eds., Strukturierung und Datierung der hethitischen Archäologie: Voraussetzungen—Probleme—Neue Ansätze; Internationaler Workshop Istanbul, 26–27. November 2004, 349–63. Istanbul: Ege Yayinlari. Gupta, A., and J. Ferguson. 1992. “Beyond ‘Culture’: Space, Identity, and the Politics of Difference.” Cultural Anthropology 7:6–23. Gurney, O. R. 1964. The Hittites. 2nd ed. Harmondsworth, Middlesex: Penguin Books. Gürtekin-Demir, R. G. 2002. “Lydian Painted Pottery at Daskyleion.” Anatolian Studies 52:111–43. Güterbock, H. G. 1983a. “The Ahhiyawa Problem Reconsidered.” American Journal of Archaeology 87:133–38. Güterbock, H. G. 1983b. “Hittite Historiography: A Survey.” In H. Tadmor and M. Weinfeld, eds., History, Historiography, and Interpretation, 21–31. Jerusalem: Magnes. Guzowska, M., and R. Becks. 2005. “Who Was Weaving at Troia? On the Aegean Style Loomweights in Troia VI and VIIa.” In R. Laffineur and E. Greco, eds., Emporia: Aegeans in the Central and Eastern Mediterranean, 279–86. Aegaeum 25. Liège: Université de Liège. Hall, E. 1989. Inventing the Barbarian. Oxford: Clarendon. Hall, J. M. 1997. Ethnic Identity in Greek Antiquity. Cambridge: Cambridge University Press. Hall, J. M. 2002. Hellenicity: Between Ethnicity and Culture. Chicago and London: University of Chicago Press. Hanfmann, J. M. A. 1978. “Lydian Relations with Ionia and Persia.” In E. Akurgal, ed., Proceedings of the Xth International Congress on Classical Archaeology, 25–35. Ankara: Turk Tarih Kurumu Basimevi. Hannerz, U. 1980. Exploring the City: Enquiries towards an Urban Anthropology. New York: Columbia University Press. Hansen, J. N., and J. N. Postgate. 1999. “The Bronze to Iron Age Transition atPage 252 → Kilise Tepe.” In A. Çilingiro lu and R. J. Matthews, eds., Anatolian Iron Ages 4,111–19. Anatolian Studies 49. London: British Institute of Archaeology at Ankara. Hartog, F. 1988. The Mirror of Herodotus: The Representation of the Other in the Writing of History. Translated by J. Lloyd. Berkeley: University of California Press. Hartog, F. 2001. Memories of Odysseus. Translated by J. Lloyd. Chicago: University of Chicago Press. Hawkes, C. 1954. “Archaeological Theory and Method: Some Suggestions from the Old World.” American Anthropologist 56:155–68. Hawkins, J. D. 1995. “The Political Geography of North Syria and South-East Anatolia in the Neo-Assyrian Period.” In M. Liverani, ed., Neo-Assyrian Geography, 87–101. Rome: Università di Roma “La Sapienza.” Hawkins, J. D. 1998. “Tarkasnawa King of Mira, ‘Tarkondemos,’ Bogaköy Sealings, and Karabel.” Anatolian Studies 48:1–31. Hawkins, J. D. 2002. “Anatolia: The End of the Hittite Empire and After.” In E. A. Braun-Holzinger and H. Mattäus, eds., Die nahöstlichen Kulturen und Griechenland an der Wende vom 2. aum 1. Jarhtausend v. Chr., 143–51. Mainz: Bibliopolis. Hawkins, J. D. 2009. “The Arzawa Letters in Recent Perspective.” British Museum Studies in Ancient Egypt and

Sudan 14:73–83. Hawkins, J. D., and D. Easton. 1996. “A Hieroglyphic Seal from Troia.” Studia Troica 6:111–18. Hayden, B. 2001. “Fabulous Feasts: A Prolegomenon to the Importance of Feasting.” In M. Dietler and B. Hayden, eds., Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, 23–64. Washington, DC: Smithsonian Institution Press. Hegmon, M. 1992. “Archaeological Research on Style.” Annual Review of Anthropology 21:517–36. Heidegger, M. [1975] 1927. Die Grundprobleme der Phänomenologie. Edited by F.-W. Herrmann. Frankfurt: Vittorio Klosterman. Heinhold-Kramer, S. 1977. Arzawa: Untersuchungen zu seiner Geschichte nachden hethitischen Quellen. Heidelberg: C. Winter. Heltzer, M. 1988. “Trade Relations between Ugarit and Crete.” Minos 23:7–13. Henrickson, R. C. 1995. “Hittite Pottery and Pooters: The View from Late Bronze Age Gordion.” Biblical Archaeologist 58:82–90. Herda, A. 2006. “Panionion-Melia, Mykalessos-Mykale, Pereus und Medusa: Überlegungen zur Besiedlungsgeschichte der Mykale in der frühen Eisenzeit.” Istanbuler Mitteilungen 56:43–102. Heuck Allen, S. 1991. “Late Bronze Age Grey Wares in Cyprus.” In J. A. Barlow, D. L. Bolger and B. Kling, eds., Cypriot Ceramics: Reading the Prehistoric Record, 151–67. Philadelphia: University Museum of Archaeology and Anthropology, University of Pennsylvania. Hindle, T. 2005. “Looking to Europe.” Economist, March 17. Hodder, I. 1982. Symbols in Action. Cambridge: Cambridge University Press. Hodder, I. 1989. The Meaning of Things: Material Culture and Symbolic Expression. London: Unwin Hyman. Page 253 → Hodder, I., and S. Hutson. 2003. Reading the Past: Current Approaches to Interpretation in Archaeology. Cambridge: Cambridge University Press. Hodder, I., and C. Orton. 1976. Spatial Analysis in Archaeology. Cambridge: Cambridge University Press. Hodges, R. 1987. “Spatial Models, Anthropology, and Archaeology.” In J. M. Wagstaff, ed., Landscape and Cultures, 118–33. Oxford: Blackwell. Hodos, T. 2006. Local Response to Colonization in the Iron Age Mediterranean. London: Routledge. Hogarth, D. G. 1909. Ionia and the East. Oxford: Clarendon. Hogarth, D. G. 1954. “Lydia and Ionia.” In J. B. Bury, S. A. Cook, and F. E. Adcock, eds., The Cambridge Ancient History, vol. 3, The Assyrian Empire, 501–26. Cambridge: Cambridge University Press. Hoggett, P. 1997. “Contested Communities.” In P. Hoggett, ed., Contested Communities, 3–16. Bristol: Policy Press. Hope Simpson, R. 2003. “The Dodecanese and the Ahhiyawa Question.” Annual of the British School at Athens 98:203–37.

Hornblower, S. 1983. The Greek World, 479–323 B.C. London: Routledge. Huxley, G. L. 1960. Achaeans and Hittites. Oxford: Oxford University Press. Hyland, A. 2003. The Horse in the Ancient World. Stroud, Gloucestershire: Sutton. Impey, O. R. 1977. Chinoiserie: The Impact of Oriental Styles on Western Art and Decoration. Oxford: Oxford University Press. Insoll, T., ed. 2006. The Archaeology of Identities: A Reader. London: Routledge. Isbell, W. H. 2000. “What We Should Be Studying: The ‘Imagined Community’ and the ‘Natural Community.’” In M. A. Canuto and J. Yaeger, eds., The Archaeology of Communities: A New World Perspective, 243–66. London: Routledge. Ivantchik, A. I. 1993. Les Cimmériens au Proche-Orient. Freiburg: Orbis Biblicus et Orientalis. Jacobson, D. 1993. Chinoiserie. London: Phaidon. Jenkins, R. 2004. Social Identity. New York: Routledge. Johnson, A. W., and T. Earle. 1987. The Evolution of Human Societies: From Foraging Group to Agrarian State. Stanford: Stanford University Press. Johnson, M. 1999. Archaeological Theory: An Introduction. Oxford: Blackwell. Johnson, R. 2006. Notting Hell. London: Fig Tree. Johnston, R. 1970. “Pottery Practices at Gordion, Central Anatolia, during the 8th to 6th Centuries B.C.: An Analyzing and Synthesizing Study.” PhD diss., Pennsylvania State University. Jones, R. E., S. T. Levi, and M. Bettelli. 2005. “Mycenaean Pottery in the Central Mediterranean: Imports, Imitations, and Derivatives.” In R. Laffineur and E. Greco, eds., Emporia: Aegeans in the Central and Eastern Mediterranean, 439–545. Aegaeum 25. Liège: Université de Liège. Jones, S. 1997. The Archaeology of Ethnicity: Constructing Identities in the Past and the Present. London: Routledge. Joukowsky, M. S. 1986. Prehistoric Aphrodisias: An Account of the Excavations and Artifact Studies. Louvain-laNeuve: Institut supérieur d'archéologie et d'histoire de l'art, Collège Érasme. Page 254 → Judah, T. 2002. Kosovo: War and Revenge. New Haven: Yale University Press. Kadish, B., and K. T. Erim. 1969. “Excavation of Prehistoric Remains at Aphrodisias, 1967.” American Journal of Archaeology 73:49–65. Karageorghis, V. 1981. Ancient Cyprus: 7,000 Years of Art and Archaeology. Baton Rouge: Louisiana State University Press. Karageorghis, V. 2001. “Why White Slip?” In V. Karageorghis, ed., The White Slip Ware of Late Bronze Age Cyprus, 9–13. Vienna: Verlag der Österreichschen Akademie der Wissenschaften. Karageorghis, V., ed. 1994. Cyprus in the 11th Century B.C. Nicosia: A. G. Leventis Foundation. Karp, D. 1997. “Community Justice: Six Challenges.” Journal of Community Psychology 27:751–69.

Katz, P. 1994. The New Urbanism: Toward an Architecture of Community. New York: McGraw-Hill. Kaya, I. 2004. Social Theory and Later Modernities. Liverpool: Liverpool University Press. Kempney, M. 2002. “Cultural Islands in the Globalizing World: Communitycum-Locality of the Ciseszyn Silesian Lutherans.” In V. Amit, ed., Realizing Community: Concepts, Social Relationships, and Sentiments, 60–83. London: Routledge. Keswani, P. S. 2005. “Death, Prestige, and Copper and Bronze Age Cyprus.” American Journal of Archaeology 109.3:341–401. Kingsley, G. T., J. B. McNeely, and J. O. Gibson. 1997. Community Building: Coming of Age. Baltimore: Development Training Institute. Klengel, H. 2002. “Problems in Hittite History, Solved and Unsolved.” In K. A. Yener and H. A. J. Hoffner, eds., Recent Developments in Hittite Archaeology and History: Papers in Memory of Hans G. Güterbock, 101–9. Winona Lake, IN: Eisenbrowns. Knapp, A. B. 2003. “The Archaeology of Community on Bronze Age Cyprus: Politko Phorades in Context.” American Journal of Archaeology 107:559–80. Knapp, A. B., and P. van Dommelen. 2008. “Past Practices: Rethinking Individuals and Agents in Archaeology.” Cambridge Archaeological Journal 19:15–34. Knappett, C. 2005. Thinking Through Material Culture. Philadelphia: University of Pennsylvania Press. Kolb, M. J., and J. E. Snead. 1997. “It's a Small World After All: Comparative Analyses of Community Organization in Archaeology.” American Antiquity 62:609–28. Konstan, D. 2001. “To Hellenikon ethnos: Ethnicity and the Construction of Ancient Greek Identity.” In I. Malkin, ed., Ancient Perceptions of Greek Ethnicity, 29–50. Cambridge, MA: Harvard University Press. Korfmann, M. 1996. “The Citadel and Lower City of Troia at Dardanelles: City of War and Peace in the Region Where Seas and Continents Meet.” In Y. Sey, ed., Habitat III: Housing and Settlement in Anatolia; A Historical Perspective. Istanbul: Tarih Vakfi Yayinlari. Kristensen, A. K. G. 1988. Who Were the Cimmerians, and Where Did They ComePage 255 → From? Sargon II, the Cimmerians, and Rusa I. Copenhagen: Det Kongelige Danske videnskabernes selskab. Kull, B. 1988. Demircihüyük. Mainz: Verlag Philipp von Zabern. Kuhrt, A. 1995. The Ancient Near East, c. 3000–330 B.C. London: Routledge. Laffineur, R., and E. Greco, eds. 2005. Emporia: Aegeans in the Central and Eastern Mediterranean. Aegaeum 25. Liège: Université de Liège. Lake, M. 2005. The EU and Turkey: A Glittering Prize or a Millstone? London: Federal Trust for Education and Research. Lamb, W. 1936. “Excavations at Kusura near Afyon Karahisar.” Archaeologia 86:1–64. Langon, P. 1994. A Better Place to Live: Reshaping the American Suburb. Amherst: University of Massachusetts Press. Layton, R. 1968. “Settlement and Community.” In P. Ucko, R. Tringham, and G. W. Dimbleby, eds., Man, Settlement, and Urbanism, 377–83. London: Duckworth.

Leach, E. 1968. “Ritual.” In S. Hugh-Jones and J. Laidlaw, eds., The Essential Edmund Leach, 1:165–73. New Haven: Yale University Press. Lekson, S. H. 2002. Review of M.A. Canuto and J. Yaeger, eds., The Archaeology of Communities: A New World Perspective. American Anthropologist 104:344–45. Lemos, I. S. 2002. The Protogeometric Aegean: The Archaeology of the Late Eleventh and Tenth Centuries BC. Oxford: Oxford University Press. Leurquin, J. L. 1986. “Prospection de la vallée du Dandalas campagne de fouilles, 1982.” In M. S. Joukowsky, Prehistoric Aphrodisias: An Account of the Excavations and Artifact Studies, 725–35. Louvain-la-Neuve: Institut supérieur d'archéologie et d'histoire de l'art, Collège Érasme. Liverani, M., ed. 1995. Neo-Assyrian Geography. Rome: Università di Roma “La Sapienza.” Lloyd, S. 1954. “Mound Surveys.” Antiquity 28:214–20. Lloyd, S. 1955. “The Citadel of an Ancient Ruler: Discoveries in Turkey.” Times (London), June 28. Lloyd, S. 1956. Early Anatolia. Harmondsworth, Middlesex: Pelican Books. Lloyd, S. 1960. “Beycesultan.” Paper presented at St Andrews University. Photocopy, British Institute of Archaeology at Ankara, London. Lloyd, S. 1972. Beycesultan. Vol. 3, part 1, Late Bronze Age Architecture. London: British Institute of Archaeology at Ankara. Lloyd, S., and J. Mellaart. 1955. “Beycesultan Excavations: First Preliminary Report.” Anatolian Studies 5:39–135. Lloyd, S., and J. Mellaart. 1956. “Beycesultan Excavations: Second Preliminary Report.” Anatolian Studies 6:101–35. Lloyd, S., and J. Mellaart. 1957. “An Early Bronze Age Shrine at Beycesultan.” Anatolian Studies 7:27–35. Lloyd, S., and J. Mellaart. 1958. “Beycesultan Excavations: Fourth Preliminary Report.” Anatolian Studies 8:93–125. Lloyd, S., and J. Mellaart. 1962. Beycesultan. Vol. 1, The Chalcolithic and Early Bronze Age Levels. London: British Institute of Archaeology at Ankara. Page 256 → Lloyd, S., and J. Mellaart. 1965. Beycesultan. Vol. 2, Middle Bronze Age Architecture and Pottery. London: British Institute of Archaeology at Ankara. Lohmann, H. 2006. “Survey in the Mycale, 3rd Campaign: The Discovery of the Archaic Panionion.” Ara tirma Sonuçlari Toplantisi, 241–52. Lohmann, H. 2007. “The Discovery and Excavation of the Archaic Panionion in the Mycale (Dilek Daglari).” Kazi Sonuçlari Toplantisi 28.2:575–90. Lovell, N. 1998. Locality and Belonging. London: Routledge. Lynch, K. 1960. The Image of the City. Cambridge, MA: MIT Press. Ma, J. 1999. Antiochos III and the Cities of Western Asia Minor. Oxford: Oxford University Press.

Mac Sweeney, N. 2009. “Beyond Ethnicity: The Overlooked Diversity of Group Identities.” Journal of Mediterranean Archaeology 22, no. 1:101–26. Mac Sweeney, N. 2010. “Hittites and Arzawans: A View from Western Anatolia.” Anatolian Studies 60:7–24. Mai Sims, J. 2008. Empowering Individuals and Creating Community: Thai Perspectives of Life in Britain. London: Runnymede Trust. Malkin, I. 1985. “What's in a Name? The Eponymous Founders of Greek Colonies.” Athenaeum 63:115–30. Malkki, L. 1992. “National Geographic: The Rooting of Peoples and the Territorialization of National Identity among Scholars and Refugees.” Cultural Anthropology 7:24–44. Mango, A. 2004. The Turks Today. London: John Murray. Marchese, R. T. 1976. “Report of West Acropolis Excavations at Aphrodisias.” American Journal of Archaeology 80:393–413. Marchese, R. T. 1986. The Lower Maeander Flood Plain: A Regional Settlement Study. BAR International Series, 292. Oxford: Archaeopress. Masson, O. 1978. Carian Inscriptions from North Saqqara and Buhen. London: Egypt Exploration Society. Matthews, R. J. 2000. “Hittites and ‘Barbarians’ in the Late Bronze Age: Regional Survey in Northern Turkey.” Archaeology International 3:32–35. Matthews, R. J. 2003. The Archaeology of Mesopotamia: Theories and Approaches. London: Routledge. Mayne, R. 1993. The Street Photographs of Roger Mayne : An Exhibition Held in the Photo Gallery of the Henry Cole Wing, from 26 February to 31 May 1986. London: Zelda Cheatle. McGaha, M. 2004. “The Poetry of Defiance: Snow.” Los Angeles Times, August 15. Mcqueen, J. G. 1968. “Geography and History in Western Asia Minor in the Second Millennium B.C.” Anatolian Studies 18:169–85. Mcqueen, J. G. 1996. The Hittites and Their Contemporaries in Asia Minor. 2nd ed. London: Thames and Hudson. Mee, C. 1998. “Anatolia and the Aegean in the Late Bronze Age.” In E. H. Cline and D. R. Harris, eds., The Aegean and the Orient in the Second Millennium, 137–48. Aegaeum 18. Liège: Université de Liège. Melchert, H. C. 2004a. “Carian.” In R. D. Woodard, ed., The Cambridge Encyclopedia of the World's Ancient Languages, 609–13. Cambridge: Cambridge University Press. Page 257 → Melchert, H. C. 2004b. “Lydian.” In R. D. Woodward, ed., The Cambridge Encyclopedia of the World's Ancient Languages, 601–8. Cambridge: Cambridge University Press. Mellaart, J. 1954. “Preliminary Report on a Survey of Pre-classical Remains in Southern Turkey.” Anatolian Studies 4:175–240. Mellaart, J. 1955. “Iron Age Pottery from Southern Anatolia.” Belleten 49:115–36. Mellaart, J. 1970. “The Second Millennium Chronology of Beycesultan.” Anatolian Studies 20:55–67.

Mellaart, J. 1982. “The Political Geography of Western Anatolia during the Late Bronze Age: Who Lived Where? ” Archiv für Orientforschung Beiheft 19:372–77. Mellaart, J. 1999. “Beycesultan.” In R. J. Matthews, ed., Ancient Anatolia: Fifty Years' Work by the British Institute of Archaeology at Ankara, 61–68. Oxford: British Institute of Archaeology at Ankara. Mellaart, J., and A. Murray. 1995. Beycesultan. Vol. 3, part 2, Late Bronze Age and Phrygian Pottery and Middle and Late Bronze Age Small Objects. London: British Institute of Archaeology at Ankara. Meriç, R. 2003. “Excavations at Bademgedigi Tepe (Puranda), 1999–2002: A Preliminary Report.” Istanbul Mitteilungen 53:79–97. Meriç, R., and P. Mountjoy. 2002. “Mycenaean Pottery from Bademgedigi Tepe (Puranda) in Ionia: A Preliminary Report.” Istanbul Mitteilungen 52:79–98. Meskell, L. 2001. “Archaeologies of Identity.” In I. Hodder, ed., Archaeological Theory Today, 187–213. Cambridge: Cambridge University Press. Meskell, L. 2006. “Archaeologies of Identity.” In T. Insoll, ed., The Archaeology of Identities: A Reader, 23–43. London: Routledge. Mewett, P. G. 1982. “Associational Categories and the Social Location of Relationships in a Lewis Crofting Community.” In A. P. Cohen, ed., Belonging: Identity and Social Organisation in British Culture, 101–30. Manchester: Manchester University Press. Mierse, W. E. 1986. “Lydian Pottery at Aphrodisias.” In M. S. Joukowsky, Prehistoric Aphrodisias: An Account of the Excavations and Artifact Studies, 413–24. Louvain-la-Neuve: Institut supérieur d'archéologie et d'histoire de l'art, Collège Érasme. Miller, M. C. 1997. Athens and Persia in the Fifth Century BC: A Study in Cultural Receptivity. Cambridge: Cambridge University Press. Mitchell, R. 2005. Country Life and the Church: The Significance of the Christian Church in an Australian Rural Community. Canberra: Glenlea. Moorey, P. R. S. 1986. “The Emergence of the Light, Horse-Drawn Chariot in the Near East, c. 2000–1500 B.C.” World Archaeology 18:196–215. Mora, C. 2003. “On Some Clauses in the Kurunta Treaty and the Political Scenery at the End of the Hittite Empire.” In G. Beckman, R. Beal, and G. McMahon, eds., Hittite Studies in Honor of Harry A. Hoffner Jr., 289–96. Winona Lake, IN: Eisenbrowns. Moran, W. L. 1992. The Amarna Letters. Baltimore: Johns Hopkins University Press. Morris, B. 1987. Anthropological Studies of Religion: An Introductory Text. Cambridge: Cambridge University Press. Page 258 → Mountjoy, P. 1993. Mycenaean Pottery: An Introduction. Oxford: School of Archaeology, University of Oxford. Mountjoy, P. 1998. “The East Aegean-West Anatolian Interface in the Late Bronze Age: Mycenaeans and the Kingdom of Ahhiyawa.” Anatolian Studies 48:33–67. Mountjoy, P. 1999. Regional Mycenaean Decorated Pottery. Rahden, Westphalia: Deutsches Archäologisches Institut.

Myres, J. L. 1930. Who Were the Greeks? Berkeley: University of California Press. Niemeier, W.-D. 1999. “Mycenaeans and Hittites in War in Asia Minor.” In R. Laffineur, ed., Polemos: Le contexte guerrier en Égée à l'Âge du Bronze, 141–55. Aegaeum 19. Liège: Université de Liège. Osborne, R. 1987. Classical Landscape with Figures: The Ancient Greek City and Its Countryside. London: Sheridan House. Osborne, R. 1996. Greece in the Making, 1299–479 BC. London: Routledge. Osborne, R. 2007. “Is Archaeology Equal to Equality?” World Archaeology 39:143–50. Ottino, A. 1998. “Origin and Ritual Exchange as Transformative Belonging in the Balinese Temple.” In N. Lovell, ed., Locality and Belonging, 103–24. London: Routledge. Özdogan, M. 2007. “Amidst Mesopotamia-centric and Euro-centric Approaches: The Changing Role of the Anatolian Peninsula between the East and the West.” In A. Irving and A. Greaves, eds., Transanatolia: Examining Turkey as a Bridge between East and West, 17–24. Anatolian Studies 57. London: British Institute of Archaeology at Ankara. Parekh, B. 2000. The Future of Multi-ethnic Britain. London: Profile Books. Park, R. E., E. W. Burgess, and R. D. MacKenzie. 1925. The City. Chicago: University of Chicago Press. Patterson, L. E. 2010. Kinship Myth in Ancient Greece. Austin: University of Texas Press. Pauketat, T. R. 2000. “Politicization and Community in the Pre-Columbian Mississippi Valley.” In M. A. Canuto and J. Yaeger, eds., The Archaeology of Communities: A New World Perspective, 16–43. London: Routledge. Pavúk, P. 2005. “Aegeans and Anatolians. A Trojan Perspective.” In R. Laffineur and E. Greco, eds., Emporia: Aegeans in the Central and Eastern Mediterranean, 269–77. Aegaeum 25. Liège: Université de Liège. Pearson, L. 1934. Early Ionian Historians. Oxford: Clarendon. Pedley, J. G. 1974. “Carians in Sardis.” Journal of Hellenic Studies 94:96–99. Peters, J. 1992. “Near-Sight and Far-Sight: Media, Place, and Culture.” In R. Rouse, J. Ferguson, and A. Gupta, eds., Culture, Power, Place: Explorations in Critical Anthropology. Boulder, CO: Westview. Peterson, C. E., and R. D. Drennan. 2005. “Communities, Settlements, Sites, and Surveys: Regional-Scale Analysis of Prehistoric Human Interaction.” American Antiquity 70:5–30. Ponti, G. 1996. “Ancient Quarrying at Aphrodisias in the Light of the Geological Configuration.” In C. Roueché and R. R. R. Smith, eds., Aphrodisias Papers, vol. 3, The Setting and Quarries, Mythological and Other Sculptural Decoration,Page 259 → Architectural Development, Portico of Tiberius, and Tetrapylon, 105–10. Journal of Roman Archaeology Supplementary Series, no. 20. Ann Arbor: Journal of Roman Archaeology. Popham, M., and L. H. Sackett. 1980. Lefkandi I: The Iron Age. London: Thames and Hudson. Postgate, J. N. 2007. “The Ceramics of Centralisation and Dissolution: A Case Study from Rough Cilicia.” In A. Irving and A. Greaves, eds., Transanatolia: Examining Turkey as a Bridge between East and West, 141–50. Anatolian Studies 57. London: British Institute of Archaeology at Ankara. Postgate, J. N., and D. Thomas. 2008. Kilise Tepe: Excavations, 1994–98. Cambridge: McDonald Institute, with the British Institute of Archaeology at Ankara. Putnam, R. 2000. Bowling Alone: The Collapse and Revival of the American Community. New York: Simon and

Schuster. Ramage, A. 1994. “Early Iron Age Sardis and Its Neighbours.” In A. Çilingiro lu and D. French, eds., Anatolian Iron Ages 3, 163–72. British Institute of Archaeology at Ankara Monograph. Ankara: British Institute of Archaeology at Ankara. Ramage, A., and P. Craddock. 2000. King Croesus' Gold: Excavations at Sardis and the History of Gold Refining. London: British Museum Press. Reese, D. S. 1986. “Shells at Aphrodisias.” In M. S. Joukowsky, Prehistoric Aphrodisias: An Account of the Excavations and Artifact Studies, 191–96. Louvain-la-Neuve: Institut supérieur d'archéologie et d'histoire de l'art, Collège Érasme. Reeves, L. 2002. “Aegean and Anatolian Bronze Age Metal Vessels: A Social Perspective.” PhD diss., Institute of Archaeology, University College London. Relph, E. 1993. “Modernity and the Reclamation of Place.” In D. Seamon, ed., Dwelling, Seeing, and Designing. Albany: State University of New York Press. Renfrew, C., and P. Bahn. 1991. Archaeology: Theories, Methods, and Practice. London: Thames and Hudson. Rice, P. M. 1987. Pottery Analysis: A Sourcebook. Chicago and London: University of Chicago Press. Rizio, A. 2005. “Vivara: An ‘International’ Port in the Bronze Age.” In R. Laffineur and E. Greco, eds., Emporia: Aegeans in the Central and Eastern Mediterranean, 623–27. Aegaeum 25. Liège: Université de Liège. Robb, J. 2001. “Island Identities: Ritual, Travel, and the Creation of Difference in Neolithic Malta.” European Journal of Archaeology 4, no. 2:175–202. Robert, L. 1954. La Carie: Histoire et Géographie Historique avec la receil des inscriptions antiques. Vol. 2, Le plateau de Tabai et ses environs. Paris: Adrien-Maisonneuve. Robert, L. 1962. Villes d'Asie Mineure: Études de géographie ancienne. Paris: E. de Boccard. Roberts, R. G. 2009. “Identity, Choice, and the Year 8 Reliefs of Ramesses III at Medinet Habu.” In C. Bachhuber and R. G. Roberts, eds., Forces of Transformation: The End of the Bronze Age in the Mediterranean, 60–68. Oxford: Oxbow. Page 260 → Robertson, R. 1995. “Glocalization: Time-Space and Homogeneity-Heterogeneity.” In R. Robertson, S. Lash, and M. Featherstone, eds., Global Modernities, 25–44. London: Thousand Oaks; New Delhi: Sage. Rockwell, P. 1996. “The Marble Quarries: A Preliminary Survey.” In C. Roueché and R. R. R. Smith, eds., Aphrodisias Papers, vol. 3, The Setting and Quarries, Mythological and Other Sculptural Decoration, Architectural Development, Portico of Tiberius, and Tetrapylon, 81–103. Journal of Roman Archaeology Supplementary Series, no. 20. Ann Arbor: Journal of Roman Archaeology. Rohrer, T. 2007. “The Body in Space: Embodiment, Experientialism, and Linguistic Conceptualisation.” In T. Ziemke, J. Zlatev, R. Frank, and R. Derven, eds., Body, Language, and Mind, 339–78. Berlin: Mouton de Gruyter. Roosevelt, C. 2009. The Archaeology of Lydia: From Gyges to Alexander. Cambridge: Cambridge University Press. Rumelli, B. 2008. “Negotiating Europe: EU-Turkey Relations from an Identity Perspective.” Insight Turkey

10:97–110. Rykwert, J. 1976. The Idea of a Town: The Anthropology of Urban Form in Rome, Italy and the Ancient World. Princeton, NJ: Princeton University Press. Sack, R. 1992. Place, Modernity, and the Consumer's World. Baltimore: John Hopkins University Press. Sagona, A., and P. Zimansky. 2009. Ancient Turkey. London: Routledge. Said, E. W. 1978. Orientalism. London: Routledge and Kegan Paul. Sams, G. K. 1974. “Phrygian Painted Animals: Anatolian Orientalizing Art.” Anatolian Studies 24:169–96. Sams, G. K. 1993. “Gordion and the Near East in the Early Phrygian Period.” In M. J. Mellink, E. Porada, and T. Ôzgûç, eds., Aspects of Art and Iconography: Anatolia and Its Neighbours; Studies in Honor of Nimet Ôzgûç, 549–55. Ankara: Türk Tarih Kurumu Basimevi. Sams, G. K. 1994. The Early Phrygian Pottery. Philadelphia: University Museum, University of Pennsylvania. Sams, G. K., and I. Temizsoy. 2000. Gordion Museum. Ankara: Türk Tarih Kurumu Basimevi. Schofield, L., and W. A. Parkinson. 1994. “Of Helmets and Heretics: A Possible Egyptian Representation of Mycenaean Warriors on a Papyrus from El-Amarna.” Annual of the British School at Athens 89:157–70. Schoop, U.-D. 2003. “Pottery Traditions of the Later Hittite Empire: Problems of Definition.” In B. Fischer, H. Genz, É. Jean, and K. Köro lu, eds., Identifying Changes: The Transition from Bronze to Iron Ages in Anatolia and the Neighbouring Regions, 167–78. Istanbul: Ege Yayinlari. Schorr, L. B. 1997. Common Purpose: Strengthening Families and Neighbourhoods to Rebuild America. New York: Anchor Books. Schwart, G. M., and S. E. Falconer, eds. 1994. Archaeological View from the Countryside: Village Communities in Early Complex Societies. Washington, DC: Smithsonian Institution Press. Seeher, J. 2005. “Überlegungen zur Beziehung zwischen dem hethitischen Kernreich und der Westküste Anatoliens im 2. Jahrtausend v. Chr.” In B. Horejs, R. Jung, E. Kaiser, and B. Terzan, eds., InterpretationsraumPage 261 → Bronzezeit: Bernhard Hänsel von seinen Schülern gewidmet, 33–44. Bonn: Rudolf Habelt. Seeley, J. R., R. A. Sim, and E. W. Loosley. 1956. Crestwood Heights. New York: Basic Books. Shanks, M., and C. Tilley. 1987. Social Theory and Archaeology. Cambridge: Cambridge University Press. Shennan, S., ed. 1994. Archaeological Approaches to Cultural Identity. London: Unwin Hyman. Sherratt, A., and S. Sherratt. 1991. “From Luxuries to Commodities: The Nature of Bronze Age Trading Systems.” In N. H. Gale, ed., Bronze Age Trade in the Mediterranean, 351–86. Jonsered: Paul Åströms Förlag. Sherratt A., and S. Sherratt. 1993. “The Growth of the Mediterranean Economy in the Early First Millennium BC.” World Archaeology 24:361–78. Sherratt, S. 1994. “Commerce, Iron, and Ideology: Metallurgical Innovation in 12th-11th Century Cyprus.” In V. Karageorghis, ed., Cyprus in the 11th Century B.C., 60–107. Nicosia: A. G. Leventis Foundation. Sherratt, S. 1999. “E pur si muove: Pots, Markets, and Values in the Second Millennium Mediterranean.” In J. P. Crielaard, V. Stissi, and G. J. van Wijngaarden, eds., The Complex Past of Pottery: Production, Circulation, and

Consumption of Mycenaean and Greek Pottery (Sixteenth to Early Fifth Centuries B.C.), 163–211. Amsterdam: J. C. Gieben. Sherratt, S. 2003. “The Mediterranean Economy: ‘Globalization’ at the End of the Second Millennium B.C.E.” In W. G. Dever and S. Gitin, eds., Symbiosis, Symbolism, and the Power of the Past, 37–62. Winona Lake, IN: Eisenbrowns. Shipley, G. 2000. The Greek World after Alexander, 323–30 BC. London: Routledge. Simmel, G. 1903. Die Grosstädte und das Geistesleben. Dresden: Petermann. Simons, W. 2001. Cities of Ladies: Beguine Communities in the Medieval Low Countries, 1200–1565. Philadelphia: University of Pennsylvania Press. im ek, C. 2007. “2005 Yili Laodikeia Kazisi Çali malari.” Kazi Sonuçlari Toplantisi 28:455–78. Singer, I. 1983. “Western Anatolia in the Thirteenth Century B.C. according to the Hittite Sources.” Anatolian Studies 33:205–17. Singer, I. 2008. “Purple-Dyers in Lazpa.” In B. J. Collins, M. R. Bachvarova, and I. C. Rutherford, eds., Anatolian Interfaces: Hittites, Greeks, and Their Neighbours, 21–44. Oxford: Oxbow. Smith, C. A. 1976. Regional Analysis. Vol. 2, Social Systems. New York: Academic Press. Smith, J. S. 2009. Art and Society in Cyprus from the Bronze Age into the Iron Age. Cambridge: Cambridge University Press. Snodgrass, A. M. 1994. “Gains, Losses, and Survivals: What We Can Infer for the Eleventh Century B.C.” In V. Karageorghis, ed., Cyprus in the 11th Century B.C., 167–75. Nicosia: A. G. Leventis Foundation. Soja, E. W. 1996. Thirdspace: Journeys to Los Angeles and Other Real-and-Imagined Places. Oxford: Blackwell. Sommer, F. 1932. Die Ahhiyava-Urkunden. Munich: Verlag der Bayerischen Akademie der Wissenschaften. Page 262 → Spawforth, A. 2001. “Shades of Greekness: A Lydian Case Study.” In I. Malkin, ed., Ancient Perceptions of Greek Ethnicity, 375–400. Cambridge, MA: Harvard University Press. Spier, J. 1983. “Prehistoric and Protohistoric Periods.” In G. M. A. Hanfmann, ed., Sardis from Prehistoric to Roman Times, 17–25. Cambridge, MA: Harvard University Press. Stacey, M. 1969. “The Myth of Community Studies.” British Journal of Sociology 20:134–47. Steele, L. 2004. Cyprus before History: From the Earliest Settlers to the End of the Bronze Age. London: Duckworth. Stos-Gale, Z. A. 2000. “An Overview of Lead Isotope Data for Provenance Studies and the Trade in Metals in the Bronze Age Mediterranean.” In C. F. E. Pare, ed., Metals Make the World Go Round: The Supply and Circulation of Metals in Bronze Age Europe, 56–69. Oxford: Oxbow. Strathern, A., and P. J. Stewart. 2000. “Creating Difference: A Contemporary Affiliation Drama in the Highlands of New Guinea.” Journal of the Royal Anthropological Institute 6:1–15. Summers, G. D. 1994. “The Recession of Mycenaean Civilization.” In I. E. S. Edwards, N. G. L. Hammond, and E. Sollberger, eds., The Cambridge Ancient History, vol. 2, part 2, The Middle East and the Aegean Region,

c.1380–1000 B.C., 338–58. Cambridge: Cambridge University Press. Thomas, J. 1996. Time, Culture, and Identity: An Interpretive Archaeology. London: Routledge. Thomas, R. 1994. “Literacy and the City-State in Archaic and Classical Greece.” In A. K. Bowman and G. Woolf, eds., Literacy and Power in the Ancient World, 33–50. Cambridge: Cambridge University Press. Thompson, D. 2007. “At the Crossroads: Prehistoric Settlement in the Maeander Valley.” In A. Irving and A. Greaves, eds., Transanatolia: Examining Turkey as a Bridge between East and West, 87–99. Anatolian Studies 57. London: British Institute of Archaeology at Ankara. Thonemann, P. J. 2006. “Maiandros: Studies in the Historical Geography of the Maeander Valley.” PhD diss., University of Oxford. Tilley, C. 1988. “Interpreting Material Culture.” In I. Hodder, ed., The Meanings of Things, 185–94. London: HarperCollins. Tilley, C. 1989. “Archaeology as Socio-political Action in the Present.” In V. Pinsky and A. Wylie, eds., Critical Traditions in Contemporary Archaeology, 104–16. Cambridge: Cambridge University Press. Tilley, C. 1990. Reading Material Culture. London: Routledge. Tilley, C. 1992. The Art of Ambiguity: Material Culture and Text. London: Routledge. Tilley, C. 1994. “Space, Place, Landscape, and Perception: Phenomenological Perspectives.” In C. Tilley, ed., A Phenomenology of Landscape, 7–34. Oxford: Berg. Tönnies, F. 1887. Gemeinschaft und Gesellschaft. Translated by C. P. Loomis, 2002. Mineola, NY: Dover Publications. Trigger, B. G. 1968. “The Determinants of Settlement Patterns.” In K. C. Chang, ed., Settlement Archaeology, 53–78. Palo Alto, CA: National Press. Page 263 → Trigger, B. G. 1989. A History of Archaeological Thought. Cambridge: Cambridge University Press. Tringham, R. 1972. “Introduction: Settlement Patterns and Urbanization.” In P. Ucko, R. Tringham, and G. W. Dimbleby, eds., Man, Settlement, and Urbanism, xix–xxviii. London: Duckworth. Tül, 1986. “Prehistoric Settlement on the Maeander Plain.” In M. S. Joukowsky, Prehistoric Aphrodisias: An Account of the Excavations and Artifact Studies, 713–24. Louvain-la-Neuve: Institut supérieur d'archéologie et d'histoire de l'art, Collège Érasme. Turner, B. S., ed. 2000. The Blackwell Companion to Social Theory. 2nd ed. Oxford: Blackwell. Vagnetti, L. 1993. “Mycenaean Pottery in Italy: Fifty Years of Study.” In C. Zerner, ed., Wace and Blegen: Pottery as Evidence of Trade in the Aegean Bronze Age, 1939–1989,143–54. Amsterdam: J. C. Gieben. Vagnetti, L. 1999. “Mycenaean Pottery in the Central Mediterranean: Imports and Local Production in Their Context.” In J. P. Crielaard, V. Stissi, and G. J. van Wijngaarden, eds., The Complex Past of Pottery: Production, Circulation, and Consumption of Mycenaean and Greek Pottery (Sixteenth to Early Fifth Centuries B.C.), 137–61. Amsterdam: J. C. Gieben. Vassileva, M. 2008. “King Midas in Southeastern Anatolia.” In B. J. Collins, M. R. Bachvarova, and I. C. Rutherford, eds., Anatolian Interfaces: Hittites, Greeks, and Their Neighbours, 165–72. Oxford: Oxbow.

van Wijngaarden, G. J. 1999. “Production, Circulation and Consumption of Mycenaean pottery (sixteenth to twelfth centuries”). In J. P. Crielaard, V. Stissi, and G. J. Van Wijngaarden, eds., The Complex Past of Pottery: Production, Circulation, and Consumption of Mycenaean and Greek Pottery (Sixteenth to Early Fifth Centuries B.C.), 21–47. Amsterdam: J. C. Gieben. van Wijngaarden, G. J. 2002. Use and Appreciation of Mycenaean Pottery in the Levant, Cyprus, and Italy (1600–1200 B.C.). Amsterdam: Amsterdam University Press. Verhoeven, M. 1999. An Archaeological Ethnography of a Neolithic Community Space: Place and Space Relations in the Burnt Village of Tell Sabi Abyal, Syria. Istanbul: Nederlands Historisch-Archaeologisch Institut. Vidal-Naquet, P. 1981. “Land and Sacrifice in the Odyssey: A Study of Religious and Mythical Meanings.” In R. Gordon, ed., Myth, Religion, and Society: Structuralist Essays by M. Detienne, L. Gernet, J.-P. Vernant, and P. Vidal-Naquet, 80–94. Cambridge: Cambridge University Press. Vogelsang-Eastwood, G. 1990. “Crescent Loomweights?” Oriens Antiquus 29:97–113. Voigt, M. M., and R. C. Henrickson. 2000. “Formation of the Phrygian State: The Early Iron Age at Gordion.” Anatolian Studies 50:37–54. Waterfield, R. 2000. The First Philosophers: The Presocratics and the Sophists. Oxford: Oxford University Press. Wernke, S. A. 2007. “Negotiating Community and Landscape in the Peruvian Andes: A Transconquest View.” American Anthropologist 109:130–52. Westgate, R., N. Fisher, and J. Whitley. 2007. Building Communities. House, Settlement,Page 264 → and Society in the Aegean and Beyond. Athens: British School at Athens. White, H. 1973. Metahistory: The Historical Imagination in Nineteenth-Century Europe. Baltimore: John Hopkins University Press. Whitelaw, T. M., and C. Morgan. 1991. “Pots and Politics: Ceramic Evidence for the Rise of the Argive State.” American Journal of Archaeology 95:79–108. Whyte, W. F. 1943. Street Corner Society: The Social Structure of an Italian Slum. Chicago: University of Chicago Press. Wiessner, P. 1984. “Reconsidering the Behavioural Basis for Style: A Case Study among the Kalhari San.” Journal of Anthropological Archaeology 3:190-234. Williams, M. W. 1956. The Sociology of an English Village: Gosforth. London: Routledge and Kegan Paul. Williamson, T., D. Imbroscio, and G. Alperowitz. 2002. Making a Place for Community. London: Routledge. Wirth, L. 1938. “Urbanism as a Way of Life.” American Journal of Sociology 44:1–24. Wright, J. C. 1994. “The Spatial Configuration of Belief: The Archaeology of Mycenaean Religion.” In R. Osborne and S. E. Alcock, eds., Placing the Gods: Sanctuaries and Sacred Space in Ancient Greece, 37–78. Oxford: Oxford University Press. Yaeger, J. 2000. “The Social Construction of Communities in the Classic Maya Countryside: Strategies of Affiliation in Western Belize.” In M. A. Canuto and J. Yaeger, eds., The Archaeology of Communities: A New World Perspective, 123–42. London: Routledge. Yaeger, J., and M. A. Canuto. 2000. “Introducing an Archaeology of Communities.” In M. A. Canuto and J. Yaeger, eds., The Archaeology of Communities: A New World Perspective, 1–15. London: Routledge.

Yale, P., J.-B. Carillet, V. Maxwell, and M. Raphael. 2005. Turkey. 9th ed. Singapore: Lonely Planet. York, P. 2006. “Are You a Notting Hillbilly? Or, Perhaps, a Shoreditch Ironist?” Independent, September 26. Young, M., and P. Willmott. [1957] 1962. Family and Kinship in East London. Rev. ed. Harmondsworth, Middlesex: Penguin Books. Zorbaugh, H. W. 1929. The Gold Coast and the Slum. Chicago: University of Chicago Press.

Page 265 → Page 266 →

Index Ahhiyawa, 66, 67, 69 Aphrodisias, 4, 58–59, 69, 78, 82, 83, 85, 88, 93, 108, 126, 140–82, 183–90 Arzawa, 66–68, 94, 110–11, 120, 136, 187, 189 Beycesultan, 4, 58–59, 68–70, 74, 82–86, 87–89, 93, 94–139, 142, 144, 145, 153, 154, 159, 164, 165, 168, 170, 171, 179, 181, 183–90 Carian, 65, 79, 92, 155, 174 “Carian Jar,” 92, 177–79 Chicago School, 12–13, 15, 24, 33 Chinoiserie, 51 chronology, 50, 83, 88, 98–100, 128, 147–49 Cohen, Anthony, 14–17 commensality, 47, 104–6, 114, 131–32, 136, 151, 170. See also dining community, archaeology of, 3–5, 22, 27–32 community, sense of, 2, 11, 14, 16, 18–21, 28–29, 34, 36–38, 42–44, 48, 63, 110–11, 118, 128, 136–38, 149, 154–55, 170, 180, 182–85, 187, 191–92 community, theories of, 9–21, 28–29 community identity, 3, 35–41, 44–58 culture history, 22–24 Cyprus, 71, 72, 73, 87, 128, 133, 164, 168, 175, 176, 188 Page 267 → Daskyleion, 73, 74, 78, 135, 142 dining, 33, 46–47, 54, 90–91, 104, 105, 106, 111, 114, 125, 132, 135, 151, 152, 156, 162–65, 167–75, 177, 180, 181, 183 Egypt, 62, 64, 68, 70, 71, 79, 80, 88, 177 enactments of community, 37, 41, 44–48, 54, 57, 89, 105–6, 109, 114, 124, 131–32, 134, 137, 151–52, 153, 155, 158, 162, 164, 166–67, 170, 174, 181, 183, 191–92 Ephesos, 68, 69, 72, 81 ethnicity, 1, 2, 13, 16, 19, 31, 32, 40, 43, 57, 82, 191 evolutionary, 12, 25, 27, 30

geographic communities, 19–21, 32 Greek, 34, 61, 63–65, 69, 74–81, 88, 90, 124, 175–76, 178, 186, 189, 190 grey ware, 69, 71, 86, 124, 126, 152, 176 Herodotus, 63, 74–75, 78, 79, 80, 81, 178, 190 Hittite, 56, 62, 64, 66–68, 70–71, 73, 94, 99, 109, 110–11, 114, 116, 118, 119–20, 126, 129, 136, 138, 154, 181–82, 186, 187, 188, 189, 190 individual identities, 37–38 Ionia/Ionians, 61, 77, 78, 79, 80, 88, 90, 145, 175–76, 189 Ionian, migration, 74–76 Karabel, 65, 68 Kilise Tepe, 87, 126, 132, 147 Knobbed Ware, 72, 87, 148, 159 Kusura, 69, 108, 153 Levant, 71, 87, 91, 96, 189 Luwian, 65, 70 Lydia/Lydians, 65, 78–81, 88, 91, 124, 146, 147, 172, 174–79, 181, 182, 188–90 Maeander, 69, 79, 81–84, 90, 93, 107–9, 140, 144 Miletus, 72, 75, 80 Mira, 66, 120, 128, 129 Mycenaean, 51, 54, 64, 66–69, 71, 87–88, 90–93, 95, 107–11, 115, 117, 126, 132–33, 136–37, 140–42, 144, 152, 162, 177, 186 Neo-Assyrian, 62, 64–65, 74, 78, 79, 177 Orientalism, 61–62 Persia, Persian, 49, 61, 63, 81 Phrygia/Phrygians, 65, 73–74, 77–78, 88, 96, 100, 124, 134–35, 137 Sardis, 68, 72–74, 78–79, 80, 84, 88, 135, 142, 147, 173–78, 190 skeuomorphism, 69, 104, 109 social rationale, 40–41, 44, 48–50, 55–57, 133–34 spatial, 20, 27, 31–33, 35–36, 41 Submycenaean, 72, 75, 100, 148

Tarhuntassa, 191, 128–29, 154 Troad, 69, 85–86, 126, 128, 155, 159, 163, 188 Troy, 68, 70–73, 94, 131, 135, 145, 148, 159, 163